Upload app_v2_entry.py with huggingface_hub

app_v2_entry.py

@@ -436,10 +436,25 @@ async def _sh(url:str='',title:str='',img:str=''):return HTMLResponse(f'<!DOCTYP
 from vtv_api import router as vtv_router
 app.include_router(vtv_router)
 
-from vtv_api import router as vtv_router
-app.include_router(vtv_router)
-
 from wc2026_scraper import(scrape_summary,scrape_fixtures,scrape_standings,scrape_stats,scrape_wc_news,scrape_road_to_wc,get_wc2026_all,scrape_history,scrape_h2h,scrape_lineups,scrape_match_detail)
+@app.get('/api/wc2026/highlights')
+def _whl():
+    """World Cup 2026 highlights from xemlaibongda.top"""
+    cache_key = 'wc2026_hl'
+    now = time.time()
+    if cache_key in _xlb_cache and now - _xlb_cache[cache_key]['t'] < _xlb_ttl:
+        return JSONResponse(_xlb_cache[cache_key]['d'])
+    try:
+        r = req.get('https://xemlaibongda.top/api/videos/highlights/the-gioi/world-cup',
+                     headers={"User-Agent": HEADERS["User-Agent"]}, timeout=15)
+        if r.status_code == 200:
+            data = r.json()
+            _xlb_cache[cache_key] = {'t': now, 'd': data}
+            return JSONResponse(data)
+    except Exception as e:
+        print(f"[WC highlights error] {e}")
+    return JSONResponse({"success": False, "data": []}, status_code=502)
+
 @app.get('/api/wc2026')
 def _w():return JSONResponse(get_wc2026_all())
 @app.get('/api/wc2026/fixtures')
@@ -504,37 +519,49 @@ def _bg():
         time.sleep(90)
 threading.Thread(target=_bg,daemon=True).start()
 
-
-# ===== VTV NAM BỘ SHORTS (fast HTML scrape, same method as Dantri/SKĐS) =====
+app.mount('/static',StaticFiles(directory=STATIC_DIR),name='vnews_static')
+# ===== VTV NAM BỘ SHORTS (Jina reader bypasses IP block) =====
 _vtv_nambo_cache = {'t': 0, 'd': []}
 _VTV_NAMBO_TTL = 1800
 
 def _scrape_vtv_nambo_shorts():
-    """Fast HTML scrape for @vtvnambo YouTube shorts."""
+    \"\"\"Scrape @vtvnambo shorts via Jina reader (bypasses YouTube IP block).\"\"\"
     now = time.time()
     if _vtv_nambo_cache['d'] and now - _vtv_nambo_cache['t'] < _VTV_NAMBO_TTL:
         return _vtv_nambo_cache['d']
     vids = []
     try:
-        url = "https://www.youtube.com/@vtvnambo/shorts"
-        r = req.get(url, headers={**HEADERS, "Accept-Language": "vi,en;q=0.8"}, timeout=15)
+        jina_url = "https://r.jina.ai/https://www.youtube.com/@vtvnambo/shorts"
+        r = req.get(jina_url, headers={"User-Agent": HEADERS["User-Agent"], "Accept": "text/plain"}, timeout=20)
         if r.status_code == 200:
-            html = r.text
-            seen_ids = []
-            for m in re.finditer(r'"videoId":"([A-Za-z0-9_-]{11})"', html):
+            text = r.text
+            found_ids = []
+            for m in re.finditer(r'youtube\.com/shorts/([A-Za-z0-9_-]{11})', text):
                 vid = m.group(1)
-                if vid in seen_ids: continue
-                seen_ids.append(vid)
-                snip = html[max(0, m.start()-900):m.start()+1600]
+                if vid in found_ids:
+                    continue
+                found_ids.append(vid)
+                # Extract nearby title
+                idx = m.start()
+                region = text[max(0, idx-300):idx+200]
                 title = ""
-                mt = re.search(r'"title":\{"runs":\[\{"text":"([^"]+)"', snip)
-                if not mt: mt = re.search(r'"accessibilityText":"([^"]+)"', snip)
-                if mt: title = html_lib.unescape(mt.group(1)).replace('\n', ' ').strip()
-                if not title: title = "VTV Nam Bộ"
-                vids.append({"title": title, "id": vid, "channel": "vtvnambo",
-                    "link": f"https://www.youtube.com/watch?v={vid}",
-                    "img": f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg", "source": "yt"})
-                if len(vids) >= 24: break
+                tm = re.search(r'### \[([^\]]+)\]', region)
+                if not tm:
+                    tm = re.search(r'\[([^\]]+)\]\(https://www\.youtube\.com/shorts', region)
+                if tm:
+                    title = tm.group(1).strip()
+                if not title or len(title) < 5:
+                    title = "VTV Nam Bộ"
+                vids.append({
+                    "title": title[:120],
+                    "link": f"https://www.youtube.com/shorts/{vid}",
+                    "img": f"https://i.ytimg.com/vi/{vid}/hqdefault.jpg",
+                    "source": "yt",
+                    "id": vid,
+                    "channel": "vtvnambo",
+                })
+                if len(vids) >= 24:
+                    break
     except Exception as e:
         print(f"[VTV Nam Bộ shorts error] {e}")
     _vtv_nambo_cache['t'] = now
@@ -545,4 +572,27 @@ def _scrape_vtv_nambo_shorts():
 def api_vtv_nambo_shorts():
     return JSONResponse(_scrape_vtv_nambo_shorts())
 
-app.mount('/static',StaticFiles(directory=STATIC_DIR),name='vnews_static')
+# ===== XEMLAIBONGDA HIGHLIGHTS PROXY =====
+_xlb_cache = {}
+_xlb_ttl = 600
+
+@app.get('/api/proxy/xlb/highlights/{category}')
+def api_xlb_highlights(category: str):
+    \"\"\"Proxy xemlaibongda highlights to avoid CORS/IP issues.\"\"\"
+    cache_key = f'xlb_hl_{category}'
+    now = time.time()
+    if cache_key in _xlb_cache and now - _xlb_cache[cache_key]['t'] < _xlb_ttl:
+        return JSONResponse(_xlb_cache[cache_key]['d'])
+    try:
+        url = f'https://xemlaibongda.top/api/videos/highlights/{category}'
+        r = req.get(url, headers={"User-Agent": HEADERS["User-Agent"]}, timeout=15)
+        if r.status_code == 200:
+            data = r.json()
+            _xlb_cache[cache_key] = {'t': now, 'd': data}
+            return JSONResponse(data)
+    except Exception as e:
+        print(f"[xlb proxy error] {e}")
+    return JSONResponse({"success": False, "data": [], "error": "fetch failed"}, status_code=502)
+@app.get('/api/vtv_nambo_shorts')
+def api_vtv_nambo_shorts():
+    return JSONResponse(_scrape_vtv_nambo_shorts())