Upload vtv_scraper.py with huggingface_hub

vtv_scraper.py

@@ -0,0 +1,156 @@
+"""
+VTV Channels Scraper
+Fetches stream URLs from hd.xemtv.net PHP endpoints for VTV1-VTV10 + VTV Cần Thơ
+The PHP endpoints return jwplayer config with fresh stream URLs (important for VTV6 which has expiring signatures)
+"""
+import requests, re, time, threading
+
+UA = {
+    "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/125.0.0.0 Safari/537.36",
+    "Accept-Language": "vi-VN,vi;q=0.9",
+    "Referer": "https://hd.xemtv.net/",
+}
+
+# Channel ID -> xemtv.net PHP endpoint mapping
+# These PHP pages return jwplayer config with fresh stream URLs
+XEMTV_PHP_ENDPOINTS = {
+    "vtv1":  "https://hd.xemtv.net/kenh/vtv1.php",
+    "vtv2":  "https://hd.xemtv.net/kenh/vtv2.php",
+    "vtv3":  "https://hd.xemtv.net/kenh/vtv3.php",
+    "vtv4":  "https://hd.xemtv.net/kenh/vtv4.php",
+    "vtv5":  "https://hd.xemtv.net/kenh/vtv5.php",
+    "vtv6":  "https://hd.xemtv.net/kenh/vtv6.php",
+    "vtv7":  "https://hd.xemtv.net/kenh/vtv7.php",
+    "vtv8":  "https://hd.xemtv.net/kenh/vtv8.php",
+    "vtv9":  "https://hd.xemtv.net/kenh/vtv9.php",
+    "vtv10": "https://hd.xemtv.net/kenh/vtv10.php",  # VTV Cần Thơ
+}
+
+# Channel display names
+CHANNEL_NAMES = {
+    "vtv1": "VTV1",
+    "vtv2": "VTV2",
+    "vtv3": "VTV3",
+    "vtv4": "VTV4",
+    "vtv5": "VTV5",
+    "vtv6": "VTV6",
+    "vtv7": "VTV7",
+    "vtv8": "VTV8",
+    "vtv9": "VTV9",
+    "vtv10": "VTV Cần Thơ",
+}
+
+# Fallback CDN streams (fptplay) — used when xemtv scraping fails
+CDN_FALLBACK = {
+    "vtv1":  "https://live.fptplay53.net/fnxch2/vtv1hd_abr.smil/chunklist.m3u8",
+    "vtv2":  "https://live.fptplay53.net/fnxch2/vtv2hd_abr.smil/chunklist.m3u8",
+    "vtv3":  "https://live.fptplay53.net/fnxch2/vtv3hd_abr.smil/chunklist.m3u8",
+    "vtv4":  "https://live.fptplay53.net/fnxch2/vtv4hd_abr.smil/chunklist.m3u8",
+    "vtv5":  "https://live-a.fptplay53.net/live/media/VTV5HD/live_hls_avc/index.m3u8",
+    "vtv7":  "https://live.fptplay53.net/fnxhd1/vtv7hd_vhls.smil/chunklist_b5000000.m3u8",
+    "vtv8":  "https://live.fptplay53.net/epzhd1/vtv8hd_vhls.smil/chunklist.m3u8",
+    "vtv9":  "https://live.fptplay53.net/fnxhd1/vtv9hd_vhls.smil/chunklist.m3u8",
+    "vtv10": "https://live.fptplay53.net/fnxch2/vtvcantho_abr.smil/chunklist.m3u8",
+    # VTV6 fallback: try canthotv as alternative
+    "vtv6":  "https://live.canthotv.vn/live/tv/chunklist.m3u8",
+}
+
+_vtv_cache = {}
+_vtv_lock = threading.Lock()
+_CACHE_TTL = 180  # 3 minutes — refresh frequently for VTV6 sign URLs
+
+
+def _cached(key):
+    with _vtv_lock:
+        if key in _vtv_cache and time.time() - _vtv_cache[key]['t'] < _CACHE_TTL:
+            return _vtv_cache[key]['d']
+    return None
+
+
+def _set_cache(key, data):
+    with _vtv_lock:
+        _vtv_cache[key] = {'t': time.time(), 'd': data}
+
+
+def extract_m3u8_from_html(html):
+    """Extract m3u8 URL from xemtv PHP page (jwplayer config)."""
+    if not html:
+        return None
+    # jwplayer config: file: 'URL'
+    m = re.search(r"file\s*:\s*['\"]([^'\"]*\.m3u8[^'\"]*)['\"]", html, re.IGNORECASE)
+    if m:
+        url = m.group(1).strip()
+        if len(url) > 20:
+            return url
+    # Generic m3u8 pattern
+    m = re.search(r"(https?://[^\s\"'<>\\]+\.m3u8[^\s\"'<>\\]*)", html, re.IGNORECASE)
+    if m:
+        url = m.group(1).strip()
+        if len(url) > 20:
+            return url
+    return None
+
+
+def fetch_vtv_stream(channel_id):
+    """Fetch m3u8 stream URL for a VTV channel by scraping xemtv.net PHP endpoint."""
+    channel_id = channel_id.lower().strip()
+
+    # Normalize name
+    name_map = {
+        'vtvct': 'vtv10', 'vtv-can-tho': 'vtv10', 'vtv can tho': 'vtv10',
+        'vtv_can_tho': 'vtv10', 'cantho': 'vtv10', 'cần thơ': 'vtv10',
+        'vietnam_vtv1': 'vtv1', 'vietnam_vtv2': 'vtv2', 'vietnam_vtv3': 'vtv3',
+        'vietnam_vtv4': 'vtv4', 'vietnam_vtv5': 'vtv5', 'vietnam_vtv6': 'vtv6',
+        'vietnam_vtv7': 'vtv7', 'vietnam_vtv8': 'vtv8', 'vietnam_vtv9': 'vtv9',
+    }
+    channel_id = name_map.get(channel_id, channel_id)
+
+    # Check cache
+    cached = _cached(channel_id)
+    if cached:
+        return cached
+
+    php_url = XEMTV_PHP_ENDPOINTS.get(channel_id)
+    if not php_url:
+        # Fallback
+        fallback = CDN_FALLBACK.get(channel_id)
+        if fallback:
+            _set_cache(channel_id, fallback)
+            return fallback
+        return None
+
+    try:
+        r = requests.get(php_url, headers=UA, timeout=15, allow_redirects=True)
+        if r.status_code == 200:
+            m3u8 = extract_m3u8_from_html(r.text)
+            if m3u8:
+                _set_cache(channel_id, m3u8)
+                return m3u8
+    except:
+        pass
+
+    # Fallback to CDN
+    fallback = CDN_FALLBACK.get(channel_id)
+    if fallback:
+        _set_cache(channel_id, fallback)
+        return fallback
+
+    return None
+
+
+def get_all_vtv_streams():
+    """Fetch all VTV channel streams. Returns list of {id, name, stream_url}."""
+    channels = []
+    for ch_id, php_url in XEMTV_PHP_ENDPOINTS.items():
+        stream_url = fetch_vtv_stream(ch_id)
+        channels.append({
+            'id': ch_id,
+            'name': CHANNEL_NAMES.get(ch_id, ch_id.upper()),
+            'stream_url': stream_url,
+        })
+    return channels
+
+
+# Legacy compatibility
+XEMTV_CHANNELS = {v: k for k, v in CHANNEL_NAMES.items()}
+CDN_STREAMS = {v: k for k, v in CDN_FALLBACK.items()}