Upload vtv_api.py with huggingface_hub

vtv_api.py

@@ -0,0 +1,463 @@
+"""
+VTV Channels API - Backend endpoints for VTV1-VTV10 + VTVPrime
+Fetches stream URLs from hd.xemtv.net PHP endpoints
+EPG data scraped from https://vtv.vn/lich-phat-song.htm
+"""
+import re, time, threading
+import requests
+from fastapi import APIRouter, Query
+from fastapi.responses import JSONResponse, Response
+from bs4 import BeautifulSoup
+from datetime import datetime, timedelta
+
+router = APIRouter()
+
+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/",
+}
+
+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",
+    "vtvprime": "https://hd.xemtv.net/kenh/vtvprime.php",
+}
+
+CHANNEL_NAMES = {
+    "vtv1":  "VTV1",
+    "vtv2":  "VTV2",
+    "vtv3":  "VTV3",
+    "vtv4":  "VTV4",
+    "vtv5":  "VTV5",
+    "vtv6":  "VTV6",
+    "vtv7":  "VTV7",
+    "vtv8":  "VTV8",
+    "vtv9":  "VTV9",
+    "vtv10": "VTV10",
+    "vtvprime": "VTVPrime",
+}
+
+VTVGO_FAILOVER = {
+    "vtv1":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv1-manifest.m3u8",
+    "vtv2":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv2-manifest.m3u8",
+    "vtv3":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv3-manifest.m3u8",
+    "vtv4":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv4-manifest.m3u8",
+    "vtv5":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv5-manifest.m3u8",
+    "vtv7":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv7-manifest.m3u8",
+    "vtv8":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv8-manifest.m3u8",
+    "vtv9":  "https://vtvgolive-failover.vtvdigital.vn/vtvgo/vtv9-manifest.m3u8",
+}
+
+XEMTV_ONLY_CHANNELS = {"vtv6", "vtv10"}
+
+FPTPLAY_URLS = {
+    "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",
+    "vtv6":  "https://live-a.fptplay53.net/live/media/vtv6/live247-hls-avc/vtv6-avc1_5600000=10000-mp4a_131600=20000.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",
+}
+
+_vtv_cache = {}
+_vtv_lock = threading.Lock()
+_CACHE_TTL = 180
+
+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):
+    if not html:
+        return None
+    m = re.search(r"file\s*:\s*['\"]([^'\"]*\.m3u8[^'\"]*)['\"]", html, re.IGNORECASE)
+    if m:
+        url = m.group(1).strip()
+        if len(url) > 20:
+            return url
+    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_xemtv_stream(channel_id):
+    php_url = XEMTV_PHP_ENDPOINTS.get(channel_id)
+    if not php_url:
+        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:
+                return m3u8
+    except:
+        pass
+    return None
+
+def fetch_fptplay_stream(channel_id):
+    url = FPTPLAY_URLS.get(channel_id)
+    if not url:
+        return None
+    try:
+        headers = {"User-Agent": UA["User-Agent"], "Referer": "https://fptplay.vn/", "Origin": "https://fptplay.vn"}
+        r = requests.get(url, headers=headers, timeout=15, allow_redirects=True)
+        if r.status_code == 200:
+            return url
+    except:
+        pass
+    return None
+
+def fetch_vtv_stream(channel_id):
+    channel_id = channel_id.lower().strip()
+    name_map = {
+        'vtvct': 'vtv10', 'vtv-can-tho': 'vtv10', 'vtv can tho': 'vtv10',
+        'vtv_can_tho': 'vtv10', 'cantho': '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)
+    cached = _cached(channel_id)
+    if cached is not None:
+        return cached
+
+    if channel_id in XEMTV_ONLY_CHANNELS:
+        xemtv_url = fetch_xemtv_stream(channel_id)
+        if xemtv_url:
+            _set_cache(channel_id, xemtv_url)
+            return xemtv_url
+        fpt_url = FPTPLAY_URLS.get(channel_id)
+        if fpt_url:
+            _set_cache(channel_id, fpt_url)
+            return fpt_url
+        _set_cache(channel_id, None)
+        return None
+
+    if channel_id in VTVGO_FAILOVER:
+        result = VTVGO_FAILOVER[channel_id]
+        _set_cache(channel_id, result)
+        return result
+
+    if channel_id == 'vtvprime':
+        xemtv_url = fetch_xemtv_stream('vtvprime')
+        if xemtv_url:
+            _set_cache(channel_id, xemtv_url)
+            return xemtv_url
+        _set_cache(channel_id, None)
+        return None
+
+    xemtv_url = fetch_xemtv_stream(channel_id)
+    if xemtv_url:
+        _set_cache(channel_id, xemtv_url)
+        return xemtv_url
+
+    fpt_url = fetch_fptplay_stream(channel_id)
+    if fpt_url:
+        _set_cache(channel_id, fpt_url)
+        return fpt_url
+
+    _set_cache(channel_id, None)
+    return None
+
+@router.get("/api/vtv/streams")
+def api_vtv_streams():
+    result = {}
+    for ch_id in CHANNEL_NAMES:
+        stream_url = fetch_vtv_stream(ch_id)
+        result[ch_id] = {"name": CHANNEL_NAMES[ch_id], "stream_url": stream_url, "status": "ok" if stream_url else "offline"}
+    return JSONResponse(result)
+
+@router.get("/api/vtv/stream/{channel_id}")
+def api_vtv_stream(channel_id: str):
+    stream_url = fetch_vtv_stream(channel_id)
+    if stream_url:
+        return JSONResponse({"stream_url": stream_url, "status": "ok"})
+    return JSONResponse({"error": "stream not found", "status": "offline"}, status_code=404)
+
+@router.get("/api/proxy/page")
+def proxy_page(url: str = Query(...)):
+    try:
+        headers = {**UA}
+        if "xemtv.net" in url:
+            headers["Referer"] = "https://hd.xemtv.net/"
+        r = requests.get(url, headers=headers, timeout=15, allow_redirects=True)
+        if r.status_code != 200:
+            return Response(status_code=502, content="upstream error")
+        return Response(content=r.text.encode("utf-8"), media_type="text/html; charset=utf-8", headers={"Access-Control-Allow-Origin": "*"})
+    except:
+        return Response(status_code=502, content="proxy error")
+
+@router.get("/api/proxy/m3u8/vtv")
+def proxy_vtv_m3u8(url: str = Query(...)):
+    try:
+        headers = {"User-Agent": UA["User-Agent"], "Accept": "*/*"}
+        if "fptplay" in url:
+            headers["Referer"] = "https://fptplay.vn/"
+            headers["Origin"] = "https://fptplay.vn"
+        elif "xemtv" in url:
+            headers["Referer"] = "https://hd.xemtv.net/"
+        r = requests.get(url, headers=headers, timeout=15, allow_redirects=True)
+        if r.status_code != 200:
+            return Response(status_code=502, content="upstream error")
+        content = r.text
+        lines = content.split('\n')
+        rewritten = []
+        base_url = url.rsplit('/', 1)[0] + '/'
+        for line in lines:
+            line = line.strip()
+            if not line or line.startswith('#'):
+                rewritten.append(line)
+            else:
+                seg_url = line
+                if not seg_url.startswith('http'):
+                    seg_url = base_url + seg_url
+                rewritten.append("/api/proxy/seg/vtv?url=" + requests.utils.quote(seg_url, safe=""))
+        return Response(content='\n'.join(rewritten).encode("utf-8"), media_type="application/vnd.apple.mpegurl", headers={"Access-Control-Allow-Origin": "*", "Cache-Control": "no-cache"})
+    except Exception as e:
+        return Response(status_code=502, content="proxy error: " + str(e))
+
+@router.get("/api/proxy/seg/vtv")
+def proxy_vtv_segment(url: str = Query(...)):
+    try:
+        headers = {"User-Agent": UA["User-Agent"], "Accept": "*/*"}
+        if "fptplay" in url:
+            headers["Referer"] = "https://fptplay.vn/"
+            headers["Origin"] = "https://fptplay.vn"
+        r = requests.get(url, headers=headers, timeout=30, allow_redirects=True)
+        if r.status_code != 200:
+            return Response(status_code=502, content="upstream error")
+        data = r.content
+        if len(data) > 188 and data[0:4] == b'\x89PNG' and data[188] == 0x47:
+            data = data[188:]
+        return Response(content=data, media_type="video/mp2t", headers={"Access-Control-Allow-Origin": "*", "Cache-Control": "public, max-age=3600"})
+    except:
+        return Response(status_code=502, content="proxy error")
+
+
+# ===== EPG from vtv.vn/lich-phat-song.htm =====
+
+_epg_cache = {}
+_epg_cache_time = 0
+_EPG_CACHE_TTL = 1800  # 30 minutes
+
+# Map vtv.vn channel IDs to our channel IDs
+VTV_CHANNEL_MAP = {
+    "vtv1": "vtv1",
+    "vtv2": "vtv2",
+    "vtv3": "vtv3",
+    "vtv4": "vtv4",
+    "vtv5": "vtv5",
+    "vtv5-tay-nam-bo": "vtv5",
+    "vtv5-tay-nguyen": "vtv5",
+    "vtv7": "vtv7",
+    "vtv8": "vtv8",
+    "vtv9": "vtv9",
+    "vtv-can-tho": "vtv10",
+}
+
+def _fetch_epg_from_vtv():
+    """Fetch EPG from https://vtv.vn/lich-phat-song.htm"""
+    global _epg_cache, _epg_cache_time
+    now_ts = time.time()
+    if _epg_cache and now_ts - _epg_cache_time < _EPG_CACHE_TTL:
+        return _epg_cache
+
+    epg_data = {}
+    try:
+        headers = {
+            "User-Agent": UA["User-Agent"],
+            "Accept-Language": "vi-VN,vi;q=0.9",
+            "Accept": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+            "Referer": "https://vtv.vn/",
+        }
+        r = requests.get("https://vtv.vn/lich-phat-song.htm", headers=headers, timeout=20)
+        if r.status_code != 200:
+            print(f"EPG vtv.vn fetch failed: {r.status_code}")
+            return epg_data
+
+        r.encoding = "utf-8"
+        soup = BeautifulSoup(r.text, "lxml")
+
+        # Get channel order from list-channel section
+        channel_order = []
+        list_channel = soup.find(class_=re.compile(r'list-channel'))
+        if list_channel:
+            for link in list_channel.find_all('a', href=re.compile(r'truyen-hinh-truc-tuyen/([^.]+)\.htm')):
+                ch_id = re.search(r'truyen-hinh-truc-tuyen/([^.]+)\.htm', link.get('href', ''))
+                if ch_id:
+                    channel_order.append(ch_id.group(1))
+
+        # If no list-channel found, try to find channel links elsewhere
+        if not channel_order:
+            for link in soup.find_all('a', href=re.compile(r'truyen-hinh-truc-tuyen/([^.]+)\.htm')):
+                ch_id = re.search(r'truyen-hinh-truc-tuyen/([^.]+)\.htm', link.get('href', ''))
+                if ch_id and ch_id.group(1) not in channel_order:
+                    channel_order.append(ch_id.group(1))
+
+        # Find all <ul class="programs"> containers
+        prog_containers = soup.find_all('ul', class_=re.compile(r'\bprograms\b'))
+
+        print(f"EPG: {len(channel_order)} channels, {len(prog_containers)} program containers")
+
+        # Map each container to channel by order
+        for i, container in enumerate(prog_containers):
+            if i >= len(channel_order):
+                break
+            vtv_ch_id = channel_order[i]
+            our_ch_id = VTV_CHANNEL_MAP.get(vtv_ch_id, vtv_ch_id)
+
+            if our_ch_id not in epg_data:
+                epg_data[our_ch_id] = []
+
+            # Parse each <li class="program"> in this container
+            for li in container.find_all('li', class_=re.compile(r'\bprogram\b')):
+                time_span = li.find('span', class_=re.compile(r'\btime\b'))
+                title_span = li.find('span', class_=re.compile(r'\btitle\b'))
+                genre_a = li.find('a', class_=re.compile(r'\bgenre\b'))
+
+                time_str = time_span.get_text(strip=True) if time_span else ""
+                # Title: prefer genre (specific program name), fallback to title (category)
+                title = ""
+                if genre_a:
+                    title = genre_a.get_text(strip=True)
+                if not title and title_span:
+                    title = title_span.get_text(strip=True)
+
+                if not time_str or not title:
+                    continue
+
+                # Parse time to datetime
+                start_dt = _parse_time(time_str)
+                if not start_dt:
+                    continue
+
+                epg_data[our_ch_id].append({
+                    "time": time_str[:5],
+                    "title": title[:80],
+                    "start_dt": start_dt,
+                })
+
+        # Sort by time and calculate end times
+        for ch_id in epg_data:
+            epg_data[ch_id].sort(key=lambda x: x.get("start_dt") or datetime.min)
+            # Remove duplicates
+            seen = set()
+            unique = []
+            for p in epg_data[ch_id]:
+                key = (p["time"], p["title"])
+                if key not in seen:
+                    seen.add(key)
+                    unique.append(p)
+            epg_data[ch_id] = unique
+
+        print(f"EPG parsed: {len(epg_data)} channels, {sum(len(v) for v in epg_data)} programmes")
+
+    except Exception as e:
+        print(f"EPG vtv.vn error: {e}")
+
+    _epg_cache = epg_data
+    _epg_cache_time = now_ts
+    return epg_data
+
+
+def _parse_time(time_str):
+    """Parse HH:MM to datetime today"""
+    if not time_str:
+        return None
+    time_str = time_str.strip().replace("h", ":").replace("H", ":")
+    m = re.search(r'(\d{1,2}):(\d{2})', time_str)
+    if m:
+        try:
+            hour, minute = int(m.group(1)), int(m.group(2))
+            now = datetime.now()
+            return now.replace(hour=hour, minute=minute, second=0, microsecond=0)
+        except:
+            pass
+    return None
+
+
+def _get_epg_for_channel(channel_id):
+    """Get EPG for a channel with current program marking."""
+    epg_data = _fetch_epg_from_vtv()
+    programmes = epg_data.get(channel_id, [])
+
+    if not programmes:
+        return []
+
+    now = datetime.now()
+    result = []
+    for i, p in enumerate(programmes):
+        start_dt = p.get("start_dt")
+        stop_dt = None
+        if i + 1 < len(programmes):
+            stop_dt = programmes[i + 1].get("start_dt")
+
+        is_now = False
+        if start_dt:
+            if stop_dt:
+                is_now = start_dt <= now < stop_dt
+            else:
+                is_now = start_dt <= now
+
+        end_time = ""
+        if stop_dt:
+            end_time = stop_dt.strftime("%H:%M")
+
+        result.append({
+            "time": p["time"],
+            "title": p["title"],
+            "end_time": end_time,
+            "now": is_now,
+        })
+
+    return result
+
+
+@router.get("/api/vtv/epg/{channel_id}")
+def api_vtv_epg(channel_id: str):
+    """Get EPG for a VTV channel from vtv.vn."""
+    channel_id = channel_id.lower().strip()
+    if channel_id not in CHANNEL_NAMES:
+        return JSONResponse({"error": "channel not found"}, status_code=404)
+    programs = _get_epg_for_channel(channel_id)
+    return JSONResponse({
+        "channel": channel_id,
+        "channel_name": CHANNEL_NAMES.get(channel_id, channel_id),
+        "date": datetime.now().strftime("%Y-%m-%d"),
+        "programs": programs,
+    })
+
+
+@router.get("/api/vtv/epg")
+def api_vtv_epg_refresh():
+    """Force refresh EPG cache."""
+    global _epg_cache, _epg_cache_time
+    _epg_cache = {}
+    _epg_cache_time = 0
+    epg_data = _fetch_epg_from_vtv()
+    return JSONResponse({
+        "status": "refreshed",
+        "channels": len(epg_data),
+        "total_programmes": sum(len(v) for v in epg_data),
+    })