fix: swap xemtivitop primary, remove Dantri/SKDS shorts from yt_scraper_fixed.py

yt_scraper_fixed.py

@@ -1,6 +1,7 @@
 """
 YouTube Shorts Scraper using yt-dlp (already installed on Space)
 Optimized for fast load with long cache + fallback
+Only VTV Nam Bộ — removed Dantri/SKDS entirely
 """
 import subprocess
 import json
@@ -11,10 +12,9 @@ import re as re_mod
 
 _cache = {}
 _lock = threading.Lock()
-CACHE_TTL = 1800  # 30 min cache - longer to reduce timeout issues
+CACHE_TTL = 1800  # 30 min cache
 
 def _get_cached(key):
-    """Get cached data if still valid"""
     with _lock:
         if key in _cache:
             entry = _cache[key]
@@ -23,12 +23,10 @@ def _get_cached(key):
     return None
 
 def _set_cached(key, data):
-    """Set cache with timestamp"""
     with _lock:
         _cache[key] = {'t': time.time(), 'd': data}
 
 def run_yt_dlp(args, timeout=45):
-    """Run yt-dlp and return parsed JSON lines - with shorter timeout"""
     try:
         result = subprocess.run(
             ["yt-dlp"] + args,
@@ -58,7 +56,7 @@ def run_yt_dlp(args, timeout=45):
         return []
 
 def get_channel_shorts_fast(channel_username, max_count=25):
-    """Get shorts fast - prioritize /shorts page only to avoid timeout"""
+    """Get shorts fast - prioritize /shorts page"""
     shorts = []
     
     url = f"https://www.youtube.com/@{channel_username}/shorts"
@@ -68,10 +66,10 @@ def get_channel_shorts_fast(channel_username, max_count=25):
         "--no-download",
         "--playlist-end", str(max_count),
         "--no-check-certificates",
-        "--quiet",  # Reduce output for speed
+        "--quiet",
         "--no-warnings",
         url
-    ], timeout=35)  # Increased timeout but still reasonable
+    ], timeout=35)
     
     seen_ids = set()
     for item in items:
@@ -94,69 +92,25 @@ def get_channel_shorts_fast(channel_username, max_count=25):
     
     return shorts
 
-def get_dantri_shorts(max_count=25):
-    """Get Dantri shorts - fast, no fallback needed, cache for 30min"""
-    cached = _get_cached('dantri_shorts_yt')
+def get_vtvnambo_shorts(max_count=50):
+    """Get shorts from VTV Nam Bộ only"""
+    cached = _get_cached('vtvnambo_shorts_yt')
     if cached is not None:
         return cached
     
-    shorts = get_channel_shorts_fast('baodantri7941', max_count)
+    shorts = get_channel_shorts_fast('vtvnambo', max_count)
     
     if shorts:
-        _set_cached('dantri_shorts_yt', shorts)
+        _set_cached('vtvnambo_shorts_yt', shorts)
         return shorts
     
     # Fallback to static list if scrape fails
-    return [
-        {"id":"Lu_iCQ5YwNM","title":"Công an lập hồ sơ xử lý người phụ nữ chửi bới tát nam tài xế ô tô ở Hà Nội","channel":"baodantri7941"},
-        {"id":"CwWvijF8BOA","title":"Chú rể Ninh Bình bật khóc nhận món quà bí mật người cha","channel":"baodantri7941"},
+    fallback = [
+        {"id":"nqlLH6chLRo","title":"Tin nóng VTV Nam Bộ | #shorts","channel":"vtvnambo"},
+        {"id":"E7Kq0v3hG6w","title":"VTV Nam Bộ - Tin tức miền Nam | #shorts","channel":"vtvnambo"},
     ]
-
-def get_skds_shorts(max_count=25):
-    """Get SKĐS shorts - fast, no fallback needed, cache for 30min"""
-    cached = _get_cached('skds_shorts_yt')
-    if cached is not None:
-        return cached
-    
-    shorts = get_channel_shorts_fast('baosuckhoedoisongboyte', max_count)
-    
-    if shorts:
-        _set_cached('skds_shorts_yt', shorts)
-        return shorts
-    
-    # Fallback to static list if scrape fails
-    return [
-        {"id":"7Pd6vZ2Lz1M","title":"Hành động ấm lòng của người đàn ông tìm kiếm 5 học sinh tử vong","channel":"baosuckhoedoisongboyte"},
-        {"id":"SlHLt_ZyPiE","title":"Xử phạt người đàn ông xóa số điện thoại cứu hộ trên cao tốc Bắc Nam","channel":"baosuckhoedoisongboyte"},
-    ]
-
-def get_dantri_skds_shorts(max_count=50):
-    """Get interleaved Dantri + SKĐS shorts - optimized with separate caching"""
-    # Get each channel's shorts separately (allows partial fallback)
-    dantri = get_dantri_shorts(max_count // 2 + 10)
-    skds = get_skds_shorts(max_count // 2 + 10)
-    
-    # Interleave them
-    result = []
-    seen = set()
-    i, j = 0, 0
-    
-    while (i < len(dantri) or j < len(skds)) and len(result) < max_count:
-        if i < len(dantri):
-            item = dantri[i]
-            if item.get('id') not in seen:
-                seen.add(item.get('id'))
-                result.append(item)
-            i += 1
-        if j < len(skds):
-            item = skds[j]
-            if item.get('id') not in seen:
-                seen.add(item.get('id'))
-                result.append(item)
-            j += 1
-    
-    return result
+    _set_cached('vtvnambo_shorts_yt', fallback)
+    return fallback
 
 # For backward compatibility
-get_vtvnambo_shorts = get_dantri_skds_shorts
-get_wc_related_shorts = get_dantri_skds_shorts
+get_wc_related_shorts = get_vtvnambo_shorts