v6.3: MyTV scraper cải tiến — dùng URL pattern từ VNEWS matches + verify URL

main.py

@@ -430,6 +430,97 @@ async def api_short_action(request:Request):
         out={"views":int(st.get("views",0)),"likes":int(st.get("likes",0)),"shares":int(st.get("shares",0)),"comments":st.get("comments",[])[:80]}
     return JSONResponse({"stats":out})
 
+# ===== MyTV.com.vn Video Scraper =====
+import unicodedata as _ud
+
+def _slugify_mytv(s):
+    """Tạo slug cho mytv URL từ tên đội"""
+    s = _ud.normalize('NFKD', s)
+    s = ''.join(c for c in s if not _ud.combining(c))
+    s = s.lower().replace(" ", "-")
+    s = re.sub(r'[^a-z0-9-]', '', s)
+    return s
+
+def scrape_mytv_highlights(limit=20):
+    """Lấy video từ mytv.com.vn cho các trận đấu đã kết thúc.
+    Vì mytv là Next.js SPA nên không scrape trực tiếp được.
+    Thay vào đó, dùng URL pattern từ các trận đấu trong VNEWS.
+    """
+    videos = []
+    seen_urls = set()
+    
+    try:
+        # Lấy trận đấu đã kết thúc trong 7 ngày qua
+        from datetime import datetime as _dt, timedelta as _td
+        now = _dt.now()
+        
+        for day_offset in range(7):
+            day = now - _td(days=day_offset)
+            ds = day.strftime("%Y-%m-%d")
+            try:
+                day_data = fetch_bongda_api(f"/api/fixtures/get-by-date?date={ds}")
+                if not day_data or len(day_data) < 100:
+                    continue
+                soup = BeautifulSoup(day_data, 'lxml')
+                for li in soup.select("li.match-detail"):
+                    status_el = li.select_one('.status .label')
+                    status_text = status_el.get_text(strip=True) if status_el else ""
+                    # Chỉ lấy trận đã kết thúc
+                    if status_text not in ('KT', 'FT', 'Finished'):
+                        continue
+                    
+                    home_el = li.select_one('.home-team .name')
+                    away_el = li.select_one('.away-team .name')
+                    home = home_el.get_text(strip=True) if home_el else ""
+                    away = away_el.get_text(strip=True) if away_el else ""
+                    
+                    if not home or not away:
+                        continue
+                    
+                    # Build mytv URL
+                    home_s = _slugify_mytv(home)
+                    away_s = _slugify_mytv(away)
+                    mytv_url = f"https://mytv.com.vn/{home_s}-vs-{away_s}"
+                    
+                    if mytv_url in seen_urls:
+                        continue
+                    seen_urls.add(mytv_url)
+                    
+                    # Verify URL exists (HEAD request, non-blocking)
+                    try:
+                        head_r = requests.head(mytv_url, headers=HEADERS, timeout=5, allow_redirects=True)
+                        if head_r.status_code == 200 and "anonymous" not in head_r.url:
+                            videos.append({
+                                "title": f"{home} vs {away}",
+                                "url": mytv_url,
+                                "source": "mytv",
+                                "type": "full_match",
+                                "home": home,
+                                "away": away,
+                                "date": ds
+                            })
+                    except:
+                        pass
+                    
+                    if len(videos) >= limit:
+                        break
+            except:
+                pass
+            
+            if len(videos) >= limit:
+                break
+    except:
+        pass
+    
+    return videos
+
+@app.get("/api/mytv/videos")
+def api_mytv_videos(limit: int = Query(default=20, le=50)):
+    """Lấy danh sách video từ mytv.com.vn"""
+    def _f():
+        return scrape_mytv_highlights(limit=limit)
+    return JSONResponse(_cached("mytv_videos", _f, ttl=300))
+
 @app.get("/api/highlights")
 def api_highlights():return JSONResponse(_cached("xemlaibongda_hl",scrape_xemlaibongda,ttl=_cache_ttl))
 @app.get("/api/highlights/leagues")