Upload match_detail_v2.py

match_detail_v2.py

@@ -1,309 +1,543 @@
+"""VNEWS — Match Detail Parser v2
+Parses bongda.com.vn match detail pages with correct selectors.
+Extracts timeline events (goals, cards, substitutions) and statistics.
 """
-Match Detail Scraper for bongda.com.vn
-"""
-import requests, re, json, time, threading
+import re
+import requests
 from bs4 import BeautifulSoup
 
-def _sp(html):
-    try:
-        return BeautifulSoup(html, 'lxml')
-    except:
-        return BeautifulSoup(html, 'html.parser')
-
-BH = {
-    "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": "application/json, text/javascript, */*; q=0.01",
-    "Referer": "https://bongda.com.vn/",
-    "X-Requested-With": "XMLHttpRequest",
-}
-HH = {
+HEADERS = {
     "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": "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
+    "Accept-Language": "vi-VN,vi;q=0.9",
     "Referer": "https://bongda.com.vn/",
 }
 
+API_HEADERS = {
+    **HEADERS,
+    "Accept": "application/json, text/javascript, */*; q=0.01",
+    "X-Requested-With": "XMLHttpRequest",
+}
+
+
+def _mk(html):
+    try:
+        return BeautifulSoup(html, 'lxml')
+    except Exception:
+        return BeautifulSoup(html, 'html.parser')
+
+
 def _cl(s):
     return re.sub(r'\s+', ' ', str(s or '')).strip()
 
-def _api(ep, params=None):
-    try:
-        url = f"https://bongda.com.vn{ep}"
-        if params:
-            url += "?" + "&".join(f"{k}={v}" for k, v in params.items())
-        r = requests.get(url, headers=BH, timeout=15)
-        if r.status_code == 200:
-            try: return r.json()
-            except: pass
-    except: pass
-    return None
-
-def _get_teams(soup):
-    info = {}
-    tel = soup.select_one('.teams')
-    if not tel:
-        return info
-    he = tel.select_one('.team.home, .home-team')
-    if he:
-        ne = he.select_one('p:not(.logo)') or he.find('p')
-        if ne: info['home_team'] = _cl(ne.get_text())
-        lo = he.select_one('img')
-        if lo: info['home_logo'] = lo.get('src', '')
-        le = he if he.name == 'a' else he.find('a')
-        if le and le.get('href'):
-            m = re.search(r'/doi-bong/(\d+)/', le['href'])
-            if m: info['home_team_id'] = m.group(1)
-    ae = tel.select_one('.team.away, .away-team')
-    if ae:
-        ne = ae.select_one('p:not(.logo)') or ae.find('p')
-        if ne: info['away_team'] = _cl(ne.get_text())
-        lo = ae.select_one('img')
-        if lo: info['away_logo'] = lo.get('src', '')
-        le = ae if ae.name == 'a' else ae.find('a')
-        if le and le.get('href'):
-            m = re.search(r'/doi-bong/(\d+)/', le['href'])
-            if m: info['away_team_id'] = m.group(1)
-    sc = tel.select_one('.score')
-    if sc:
-        parts = [_cl(p.get_text()) for p in sc.select('p')]
-        if len(parts) >= 2: info['score'] = f"{parts[0]} - {parts[1]}"
-        lb = sc.select_one('.label')
-        if lb: info['status_label'] = _cl(lb.get_text())
-    return info
-
-def _get_timeline(soup):
-    tl = []
-    el = soup.select_one('.timeline')
-    if not el: return tl
-    half = ''
-    for c in el.children:
-        if not hasattr(c, 'name') or not c.name: continue
-        t = _cl(c.get_text())
-        if not t: continue
-        if t in ['H1','H2','Hiệp 1','Hiệp 2']:
-            half = t; continue
-        m = re.match(r"(\d+'\+?\d*)", t)
-        if m:
-            tl.append({'time': m.group(1), 'text': t[m.end():].strip(), 'half': half})
-        elif len(t) > 5:
-            tl.append({'time': '', 'text': t, 'half': half})
-    return tl
-
-def _get_events(soup):
-    evts = []
-    for el in soup.select('.event'):
-        e = {}
-        cl = ' '.join(el.get('class', []))
-        e['team'] = 'home' if 'home' in cl else ('away' if 'away' in cl else '')
-        ps = [_cl(p.get_text()) for p in el.select('p')]
-        ps = [p for p in ps if p]
-        if ps: e['players'] = ps
-        tl = el.select_one('.time, .minute, span')
-        if tl: e['time'] = _cl(tl.get_text())
-        evts.append(e)
-    return evts
-
-def _get_stats(soup):
-    st = {}
-    for sel in ['.match-stats','[class*="stats"]']:
-        el = soup.select_one(sel)
-        if el and len(str(el)) > 50:
-            for row in el.select('li,tr,.stat-row'):
-                cells = row.select('td,span,p')
-                if len(cells) >= 3:
-                    lb = _cl(cells[0].get_text())
-                    if lb: st[lb] = {'home': _cl(cells[1].get_text()), 'away': _cl(cells[2].get_text())}
-            if st: break
-    return st
-
-def _get_h2h(soup):
-    h2h = {'matches': [], 'stats': {}}
-    for sel in ['.head-to-head','[class*="h2h"]']:
-        el = soup.select_one(sel)
-        if el and len(str(el)) > 50:
-            for it in el.select('li,tr,.match-item'):
-                m = {}
-                cells = it.select('td,span,p')
-                if len(cells) >= 3:
-                    m['date'] = _cl(cells[0].get_text())
-                    m['home'] = _cl(cells[1].get_text())
-                    m['score'] = _cl(cells[2].get_text())
-                if m.get('home'):
-                    if len(cells) > 3: m['away'] = _cl(cells[3].get_text())
-                    h2h['matches'].append(m)
-            if h2h['matches']: break
-    return h2h
-
-def _get_form(soup):
-    f = {'home': [], 'away': []}
-    for sel in ['.form-guide','[class*="form"]']:
-        el = soup.select_one(sel)
-        if el and len(str(el)) > 50:
-            items = el.select('li,.form-item,tr')
-            for it in items[:10]:
-                t = _cl(it.get_text())
-                if t: f['home'].append({'text': t})
-            for it in items[10:20]:
-                t = _cl(it.get_text())
-                if t: f['away'].append({'text': t})
-            break
-    return f
-
-def _get_info(soup):
+
+def _base_url(url_or_eid):
+    """Extract base match URL."""
+    if isinstance(url_or_eid, int) or url_or_eid.isdigit():
+        return f"https://bongda.com.vn/tran-dau/{int(url_or_eid)}"
+    url = url_or_eid.strip()
+    url = re.sub(r'/(preview|centre|bao-cao-nhanh|thong-ke|doi-hinh)/.*', '', url)
+    return url.rstrip('/')
+
+
+def fetch_html(url, timeout=15):
+    """Fetch HTML from URL."""
+    resp = requests.get(url, headers=HEADERS, timeout=timeout, allow_redirects=True)
+    resp.raise_for_status()
+    return resp.text
+
+
+def parse_events(sp):
+    """Parse timeline events from BeautifulSoup object.
+    
+    Structure: .events > .period > .event.home-team|away-team
+    
+    Each event:
+      .event-type > icon (goal/redcard/yellowcard/substitution)
+      .players.card|goal|subst > .event-time + player names
+      
+    Returns list of event dicts with: type, time, team, players, player_in, player_out, scorer, assist
+    """
+    events = []
+    events_div = sp.select_one('.events')
+    if not events_div:
+        return events
+
+    current_period = ''
+    for child in events_div.children:
+        if not hasattr(child, 'name') or not child.name:
+            continue
+        cls = child.get('class', [])
+        cls_str = ' '.join(cls)
+
+        if 'period' in cls_str:
+            # Period header contains h2 with period name
+            h2 = child.find('h2')
+            if h2:
+                current_period = _cl(h2.get_text())
+
+            # Events inside period
+            for ev in child.children:
+                if not hasattr(ev, 'name') or not ev.name:
+                    continue
+                ev_cls = ev.get('class', [])
+                ev_cls_str = ' '.join(ev_cls)
+                if 'event' not in ev_cls_str:
+                    continue
+
+                team = 'home' if 'home' in ev_cls_str else 'away'
+                ev_data = {
+                    'team': team,
+                    'period': current_period,
+                    'type': 'unknown',
+                    'time': '',
+                    'players': '',
+                    'player_in': '',
+                    'player_out': '',
+                    'scorer': '',
+                    'assist': '',
+                    'card_type': '',
+                    'player': '',
+                }
+
+                # Determine event type from icon in .event-type
+                type_el = ev.select_one('.event-type')
+                if type_el:
+                    # Check for specific SVG/icon patterns
+                    if type_el.select_one('[class*="redcard"]'):
+                        ev_data['type'] = 'redcard'
+                        ev_data['card_type'] = 'red'
+                    elif type_el.select_one('[class*="yellowcard"]'):
+                        ev_data['type'] = 'yellowcard'
+                        ev_data['card_type'] = 'yellow'
+                    elif type_el.select_one('[class*="goal"]'):
+                        ev_data['type'] = 'goal'
+                    elif type_el.select_one('[class*="substitution"]'):
+                        ev_data['type'] = 'substitution'
+                    else:
+                        # Check by SVG content
+                        svg_rects = type_el.select('svg rect')
+                        for rect in svg_rects:
+                            fill = rect.get('fill', '')
+                            if fill == '#E20007':
+                                ev_data['type'] = 'redcard'
+                                ev_data['card_type'] = 'red'
+                                break
+                        if ev_data['type'] == 'unknown':
+                            svg_circles = type_el.select('svg circle')
+                            for circle in svg_circles:
+                                fill = circle.get('fill', '')
+                                if fill == 'white' and circle.get('r') == '8':
+                                    ev_data['type'] = 'goal'
+                                    break
+                        if ev_data['type'] == 'unknown':
+                            if type_el.select_one('[class*="subst"]') or type_el.select('svg'):
+                                # Check for substitution SVG pattern
+                                svg = type_el.select_one('svg')
+                                if svg and ev.select_one('.players.subst'):
+                                    ev_data['type'] = 'substitution'
+
+                # Also check .players class for type hint
+                players_el = ev.select_one('.players')
+                if players_el and ev_data['type'] == 'unknown':
+                    pcls = ' '.join(players_el.get('class', []))
+                    if 'card' in pcls and ev_data['type'] == 'unknown':
+                        ev_data['type'] = 'redcard'  # default to red if card class but no type detected
+                    elif 'goal' in pcls:
+                        ev_data['type'] = 'goal'
+                    elif 'subst' in pcls:
+                        ev_data['type'] = 'substitution'
+
+                # Time from .event-time
+                if players_el:
+                    time_el = players_el.select_one('.event-time')
+                    if time_el:
+                        ev_data['time'] = _cl(time_el.get_text())
+
+                    # Player text
+                    ev_data['players'] = _cl(players_el.get_text(' ', strip=True))
+
+                    # Parse player names based on event type
+                    texts = []
+                    for d in players_el.find_all('div', recursive=False):
+                        t = _cl(d.get_text())
+                        if t and t != ev_data['time']:
+                            # Remove trailing ' if present
+                            texts.append(t)
+
+                    # Also get text from <p> elements directly in .players
+                    for p in players_el.find_all('p', recursive=False):
+                        t = _cl(p.get_text())
+                        if t and t not in texts:
+                            texts.append(t)
+
+                    if ev_data['type'] == 'substitution':
+                        if len(texts) >= 2:
+                            ev_data['player_out'] = texts[0]
+                            ev_data['player_in'] = texts[1]
+                        elif len(texts) == 1:
+                            ev_data['player_in'] = texts[0]
+                        # Also try from combined text: "time' playerOut playerIn"
+                        if not ev_data['player_in'] and ev_data['players']:
+                            text = ev_data['players'].replace(ev_data['time'], '').strip()
+                            words = text.split()
+                            if len(words) >= 4:
+                                # "46' Jose Sa Rui Silva" => out="Jose Sa", in="Rui Silva"
+                                mid = len(words) // 2
+                                ev_data['player_out'] = ' '.join(words[:mid])
+                                ev_data['player_in'] = ' '.join(words[mid:])
+
+                    elif ev_data['type'] == 'goal':
+                        if len(texts) >= 1:
+                            ev_data['scorer'] = texts[0]
+                        if len(texts) >= 2:
+                            ev_data['assist'] = texts[1]
+                        if not ev_data['scorer'] and ev_data['players']:
+                            text = ev_data['players'].replace(ev_data['time'], '').strip()
+                            words = text.split()
+                            if len(words) >= 2:
+                                ev_data['scorer'] = ' '.join(words[:2])
+                            elif len(words) == 1:
+                                ev_data['scorer'] = words[0]
+
+                    elif ev_data['type'] in ('redcard', 'yellowcard'):
+                        if texts:
+                            ev_data['player'] = ' '.join(texts)
+                        elif ev_data['players']:
+                            text = ev_data['players'].replace(ev_data['time'], '').strip()
+                            if text:
+                                ev_data['player'] = text
+
+                events.append(ev_data)
+
+    return events
+
+
+def fetch_match_detail(event_id: int) -> dict:
+    """Fetch and parse match detail by event ID."""
+    result = {"event_id": event_id, "found": False, "sections": []}
+
+    html = None
+    base = f"https://bongda.com.vn/tran-dau/{event_id}"
+    for suffix in ['/centre/', '/preview/', '/bao-cao-nhanh/']:
+        url = base + suffix
+        try:
+            html = fetch_html(url, timeout=15)
+            if html and len(html) > 1000 and '.teams' in html:
+                break
+        except Exception:
+            html = None
+            continue
+
+    if not html:
+        return result
+
+    sp = _mk(html)
     info = {}
-    mi = soup.select_one('.match-info')
-    if mi:
-        te = mi.select_one('.times,li')
-        if te: info['datetime'] = _cl(te.get_text())
-    le = soup.select_one('.league,.tournament,[class*="league"]')
-    if le: info['league'] = _cl(le.get_text())
-    return info
-
-def _scrape(url):
-    print(f"[DEBUG] _scrape: {url[:80]}", flush=True)
-    try:
-        r = requests.get(url, headers=HH, timeout=15, allow_redirects=True)
-        print(f"[DEBUG] HTTP={r.status_code}", flush=True)
-        if r.status_code != 200:
-            return False, {}
-        sp = _sp(r.text)
-        d = {}
-        
-        teams = _get_teams(sp)
-        print(f"[DEBUG] teams={teams}", flush=True)
-        if teams: d['info'] = teams
-        
-        mi = _get_info(sp)
-        if mi:
-            d.setdefault('info', {}).update(mi)
-        
-        tl = _get_timeline(sp)
-        if tl:
-            d['timeline'] = tl
-            d['commentaries_html'] = '\n'.join([f"{t.get('time','')} {t.get('text','')}" for t in tl])
-        
-        ev = _get_events(sp)
-        if ev: d['events'] = ev
+
+    tel = sp.select_one('.teams')
+    if tel:
+        he = tel.select_one('.team.home') or tel.select_one('[class*="home"]')
+        if he:
+            ne = he.select_one('p:not(.logo)') or he.find('p')
+            if ne:
+                info['home_team'] = _cl(ne.get_text())
+            lo = he.select_one('img')
+            if lo:
+                info['home_logo'] = lo.get('src', '')
         
-        st = _get_stats(sp)
-        if st:
-            d['stats_parsed'] = st
-            d['stats_html'] = str(st)
+        ae = tel.select_one('.team.away') or tel.select_one('[class*="away"]')
+        if ae:
+            ne = ae.select_one('p:not(.logo)') or ae.find('p')
+            if ne:
+                info['away_team'] = _cl(ne.get_text())
+            lo = ae.select_one('img')
+            if lo:
+                info['away_logo'] = lo.get('src', '')
         
-        h2h = _get_h2h(sp)
-        if h2h.get('matches'): d['h2h_matches'] = h2h['matches']
-        if h2h.get('stats'): d['h2h_stats'] = h2h['stats']
+        sc = tel.select_one('.score')
+        if sc:
+            parts = [_cl(p.get_text()) for p in sc.select('p')]
+            if len(parts) >= 2:
+                info['score'] = f"{parts[0]} - {parts[1]}"
+            lb = sc.select_one('.label')
+            if lb:
+                info['status_label'] = _cl(lb.get_text())
+
+    if info.get('home_team') and info.get('away_team'):
+        result['info'] = info
+        result['found'] = True
+        result['sections'].append('info')
+    else:
+        return result
+
+    mi = sp.select_one('.match-info')
+    if mi:
+        for sel_key in ['.times', 'li']:
+            el = mi.select_one(sel_key)
+            if el:
+                t = _cl(el.get_text())
+                if t:
+                    info.setdefault('datetime', t)
+                    break
+
+    events = parse_events(sp)
+    if events:
+        result['events'] = events
+        result['sections'].append('events')
+
+    # Prediction
+    pred = sp.select_one('.prediction-card')
+    if pred:
+        pred_data = {}
+        team_info = pred.select_one('.team-info')
+        if team_info:
+            teams = team_info.select('.team')
+            if len(teams) >= 2:
+                pred_data['home_name'] = _cl(teams[0].select_one('.team-name').get_text() if teams[0].select_one('.team-name') else '')
+                pred_data['away_name'] = _cl(teams[1].select_one('.team-name').get_text() if teams[1].select_one('.team-name') else '')
+                divider = team_info.select_one('.divider')
+                if divider:
+                    pred_data['result'] = _cl(divider.get_text())
+        vote_count = pred.select_one('.vote-count')
+        if vote_count:
+            pred_data['vote_count'] = _cl(vote_count.get_text())
+        result['prediction'] = pred_data
+
+    # H2H stats from API
+    try:
+        ar = requests.get(
+            f"https://bongda.com.vn/api/fixtures/h2h-stats?event_id={event_id}",
+            headers=API_HEADERS,
+            timeout=10
+        )
+        if ar.status_code == 200:
+            ad = ar.json()
+            if ad.get('status') == 'success' and ad.get('html'):
+                asp = _mk(ad['html'])
+                ast = {}
+                for row in asp.select('li, tr, .stat-row'):
+                    cells = row.select('td, span, p')
+                    if len(cells) >= 3:
+                        lb = _cl(cells[0].get_text())
+                        if lb:
+                            ast[lb] = {
+                                'home': _cl(cells[1].get_text()),
+                                'away': _cl(cells[2].get_text()),
+                            }
+                if ast:
+                    result['h2h_stats_parsed'] = ast
+                    result['sections'].append('h2h_stats')
+    except Exception:
+        pass
+
+    # H2H standings
+    h2h_data = []
+    h2h_el = sp.select_one('.h2h-standings')
+    if h2h_el:
+        rows = h2h_el.select('.ranking-table .body > tr, .ranking-table tbody tr, .leaderboard tr')
+        for row in rows:
+            cells = row.select('td')
+            if len(cells) >= 4:
+                logo = row.select_one('img')
+                name_el = row.select_one('.team-name, p.link, .name')
+                h2h_data.append({
+                    'pos': _cl(cells[0].get_text()) if cells else '',
+                    'logo': logo.get('src', '') if logo else '',
+                    'name': _cl(name_el.get_text()) if name_el else '',
+                    'played': _cl(cells[1].get_text()) if len(cells) > 1 else '',
+                    'wins': _cl(cells[2].get_text()) if len(cells) > 2 else '',
+                    'draws': _cl(cells[3].get_text()) if len(cells) > 3 else '',
+                    'losses': _cl(cells[4].get_text()) if len(cells) > 4 else '',
+                    'gf': _cl(cells[5].get_text()) if len(cells) > 5 else '',
+                    'ga': _cl(cells[6].get_text()) if len(cells) > 6 else '',
+                    'points': _cl(cells[8].get_text()) if len(cells) > 8 else '',
+                })
+        if h2h_data:
+            result['h2h_standings'] = h2h_data
+            result['sections'].append('h2h_standings')
+
+    # Recent matches
+    recent_matches = []
+    matches_list = sp.select_one('.matches-list')
+    if matches_list:
+        for item in matches_list.select('.match-detail, .match-item, li'):
+            date_el = item.select_one('.date, .time, .match-time')
+            league_el = item.select_one('.league')
+            home_el = item.select_one('.home, .team-home')
+            away_el = item.select_one('.away, .team-away')
+            score_el = item.select_one('.score, .result')
+            if home_el or away_el:
+                recent_matches.append({
+                    'date': _cl(date_el.get_text()) if date_el else '',
+                    'league': _cl(league_el.get_text()) if league_el else '',
+                    'home': _cl(home_el.get_text()) if home_el else '',
+                    'away': _cl(away_el.get_text()) if away_el else '',
+                    'score': _cl(score_el.get_text()) if score_el else 'vs',
+                })
+        if recent_matches:
+            result['recent_matches'] = recent_matches
+            result['sections'].append('recent')
+
+    return result
+
+
+def fetch_match_detail_by_url(url: str) -> dict:
+    """Fetch and parse match detail from a full bongda.com.vn URL."""
+    import os
+    eid_match = re.search(r'/tran-dau/(\d+)/', url)
+    if not eid_match:
+        return {"event_id": 0, "found": False, "error": "Cannot extract event_id from URL"}
+    
+    event_id = int(eid_match.group(1))
+    result = {"event_id": event_id, "found": False, "sections": []}
+
+    html = None
+    try:
+        html = fetch_html(url, timeout=15)
+    except Exception:
+        pass
+
+    if not html or len(html) < 1000 or '.teams' not in html:
+        return fetch_match_detail(event_id)
+
+    sp = _mk(html)
+    info = {}
+
+    tel = sp.select_one('.teams')
+    if tel:
+        he = tel.select_one('.team.home') or tel.select_one('[class*="home"]')
+        if he:
+            ne = he.select_one('p:not(.logo)') or he.find('p')
+            if ne:
+                info['home_team'] = _cl(ne.get_text())
+            lo = he.select_one('img')
+            if lo:
+                info['home_logo'] = lo.get('src', '')
         
-        if '/preview/' in url:
-            fm = _get_form(sp)
-            if fm.get('home'): d['home_form'] = fm['home']
-            if fm.get('away'): d['away_form'] = fm['away']
+        ae = tel.select_one('.team.away') or tel.select_one('[class*="away"]')
+        if ae:
+            ne = ae.select_one('p:not(.logo)') or ae.find('p')
+            if ne:
+                info['away_team'] = _cl(ne.get_text())
+            lo = ae.select_one('img')
+            if lo:
+                info['away_logo'] = lo.get('src', '')
         
-        print(f"[DEBUG] success keys={list(d.keys())}", flush=True)
-        return True, d
-    except Exception as e:
-        import traceback
-        print(f"[DEBUG] error: {e}", flush=True)
-        traceback.print_exc()
-        return False, {}
-
-def fetch_match_detail_by_url(url):
-    m = re.search(r'/tran-dau/(\d+)/', url)
-    if not m: return {"error": "Could not extract event_id", "found": False}
-    event_id = int(m.group(1))
-    res = {"event_id": event_id, "found": False, "sections": []}
-    _fetch_api(event_id, res)
-    ok, d = _scrape(url)
-    print(f"[DEBUG] by_url: ok={ok} d_keys={list(d.keys())}", flush=True)
-    if ok: _merge(res, d)
-    return res
-
-def fetch_match_detail(event_id):
-    print(f"[DEBUG] fetch_match_detail({event_id})", flush=True)
-    res = {"event_id": event_id, "found": False, "sections": []}
-    _fetch_api(event_id, res)
-    
-    for pt in ["centre", "preview"]:
-        url = f"https://bongda.com.vn/tran-dau/{event_id}/{pt}/"
-        ok, d = _scrape(url)
-        print(f"[DEBUG] {pt}: ok={ok}", flush=True)
-        if ok:
-            _merge(res, d)
-            if res.get("found"): break
-    
-    print(f"[DEBUG] final: found={res['found']} sections={res['sections']}", flush=True)
-    return res
+        sc = tel.select_one('.score')
+        if sc:
+            parts = [_cl(p.get_text()) for p in sc.select('p')]
+            if len(parts) >= 2:
+                info['score'] = f"{parts[0]} - {parts[1]}"
+            lb = sc.select_one('.label')
+            if lb:
+                info['status_label'] = _cl(lb.get_text())
 
-def _fetch_api(eid, res):
-    pm = _api("/api/event-standing/pre-match", {"event_id": eid})
-    res["pre_match"] = pm
-    res["pre_match_html"] = pm.get("html","") if pm and pm.get("status")=="success" and len(pm.get("html","").strip())>10 else ""
-    
-    hm = _api("/api/fixtures/h2h-match", {"event_id": eid})
-    res["h2h_match"] = hm
-    if hm and hm.get("status")=="success":
-        h = hm.get("html","")
-        if len(h.strip())>10:
-            res["h2h_html"] = h
-            res["sections"].append("h2h")
-    else: res["h2h_html"] = ""
-    
-    hs = _api("/api/fixtures/h2h-stats", {"event_id": eid})
-    res["h2h_stats"] = hs
-    if hs and hs.get("status")=="success":
-        h = hs.get("html","")
-        if len(h.strip())>10:
-            res["h2h_stats_html"] = h
-            res["sections"].append("h2h_stats")
-            try:
-                sp = _sp(h)
-                stats = {}
-                for row in sp.select('li,tr,.stat-row'):
-                    cells = row.select('td,span,p')
-                    if len(cells)>=3:
+    if info.get('home_team') and info.get('away_team'):
+        result['info'] = info
+        result['found'] = True
+        result['sections'].append('info')
+    else:
+        return fetch_match_detail(event_id)
+
+    mi = sp.select_one('.match-info')
+    if mi:
+        te = mi.select_one('.times, li')
+        if te:
+            info.setdefault('datetime', _cl(te.get_text()))
+
+    events = parse_events(sp)
+    if events:
+        result['events'] = events
+        result['sections'].append('events')
+
+    pred = sp.select_one('.prediction-card')
+    if pred:
+        pred_data = {}
+        team_info = pred.select_one('.team-info')
+        if team_info:
+            teams = team_info.select('.team')
+            if len(teams) >= 2:
+                pred_data['home_name'] = _cl(teams[0].select_one('.team-name').get_text() if teams[0].select_one('.team-name') else '')
+                pred_data['away_name'] = _cl(teams[1].select_one('.team-name').get_text() if teams[1].select_one('.team-name') else '')
+                divider = team_info.select_one('.divider')
+                if divider:
+                    pred_data['result'] = _cl(divider.get_text())
+        vote_count = pred.select_one('.vote-count')
+        if vote_count:
+            pred_data['vote_count'] = _cl(vote_count.get_text())
+        result['prediction'] = pred_data
+
+    try:
+        ar = requests.get(
+            f"https://bongda.com.vn/api/fixtures/h2h-stats?event_id={event_id}",
+            headers=API_HEADERS,
+            timeout=10
+        )
+        if ar.status_code == 200:
+            ad = ar.json()
+            if ad.get('status') == 'success' and ad.get('html'):
+                asp = _mk(ad['html'])
+                ast = {}
+                for row in asp.select('li, tr, .stat-row'):
+                    cells = row.select('td, span, p')
+                    if len(cells) >= 3:
                         lb = _cl(cells[0].get_text())
-                        if lb: stats[lb] = {'home': _cl(cells[1].get_text()), 'away': _cl(cells[2].get_text())}
-                if stats: res["h2h_stats_parsed"] = stats
-            except: pass
-    else: res["h2h_stats_html"] = ""
-    
-    pf = _api("/api/event-standing/player-performance", {"event_id": eid})
-    res["performance"] = pf
-    if pf and pf.get("status")=="success" and len(pf.get("html","").strip())>10:
-        res["stats_html"] = pf["html"]
-        res["sections"].append("stats")
-    else: res["stats_html"] = ""
-    
-    cm = _api("/api/fixtures/commentaries", {"event_id": eid})
-    if cm and cm.get("status")=="success" and len(cm.get("html","").strip())>10:
-        res["commentaries_html"] = cm["html"]
-        res["sections"].append("commentaries")
-    elif not res.get("commentaries_html"): res["commentaries_html"] = ""
-
-def _merge(res, d):
-    if d.get("info"):
-        res.setdefault("info", {}).update(d["info"])
-        res["found"] = True
-        if "info" not in res["sections"]: res["sections"].append("info")
-    if d.get("timeline"):
-        res["timeline"] = d["timeline"]
-        if not res.get("commentaries_html"): res["commentaries_html"] = d.get("commentaries_html","")
-        res["sections"].append("commentaries")
-    if d.get("events"):
-        res["events"] = d["events"]
-        res["sections"].append("events")
-    if d.get("stats_parsed"):
-        res["stats_parsed"] = d["stats_parsed"]
-        if not res.get("stats_html"): res["stats_html"] = d.get("stats_html","")
-        res["sections"].append("stats")
-    if d.get("h2h_matches"):
-        res["h2h"] = d["h2h_matches"]
-        res["sections"].append("h2h")
-    if d.get("h2h_stats"):
-        res["h2h_stats_parsed"] = d["h2h_stats"]
-        res["sections"].append("h2h_stats")
-    if d.get("home_form"):
-        res["home_form"] = d["home_form"]
-        res["sections"].append("home_form")
-    if d.get("away_form"):
-        res["away_form"] = d["away_form"]
-        res["sections"].append("away_form")
+                        if lb:
+                            ast[lb] = {
+                                'home': _cl(cells[1].get_text()),
+                                'away': _cl(cells[2].get_text()),
+                            }
+                if ast:
+                    result['h2h_stats_parsed'] = ast
+                    result['sections'].append('h2h_stats')
+    except Exception:
+        pass
+
+    h2h_data = []
+    h2h_el = sp.select_one('.h2h-standings')
+    if h2h_el:
+        rows = h2h_el.select('.ranking-table .body > tr, .ranking-table tbody tr, .leaderboard tr')
+        for row in rows:
+            cells = row.select('td')
+            if len(cells) >= 4:
+                logo = row.select_one('img')
+                name_el = row.select_one('.team-name, p.link, .name')
+                h2h_data.append({
+                    'pos': _cl(cells[0].get_text()) if cells else '',
+                    'logo': logo.get('src', '') if logo else '',
+                    'name': _cl(name_el.get_text()) if name_el else '',
+                    'played': _cl(cells[1].get_text()) if len(cells) > 1 else '',
+                    'wins': _cl(cells[2].get_text()) if len(cells) > 2 else '',
+                    'draws': _cl(cells[3].get_text()) if len(cells) > 3 else '',
+                    'losses': _cl(cells[4].get_text()) if len(cells) > 4 else '',
+                    'gf': _cl(cells[5].get_text()) if len(cells) > 5 else '',
+                    'ga': _cl(cells[6].get_text()) if len(cells) > 6 else '',
+                    'points': _cl(cells[8].get_text()) if len(cells) > 8 else '',
+                })
+        if h2h_data:
+            result['h2h_standings'] = h2h_data
+            result['sections'].append('h2h_standings')
+
+    recent_matches = []
+    matches_list = sp.select_one('.matches-list')
+    if matches_list:
+        for item in matches_list.select('.match-detail, .match-item, li'):
+            date_el = item.select_one('.date, .time, .match-time')
+            league_el = item.select_one('.league')
+            home_el = item.select_one('.home, .team-home')
+            away_el = item.select_one('.away, .team-away')
+            score_el = item.select_one('.score, .result')
+            if home_el or away_el:
+                recent_matches.append({
+                    'date': _cl(date_el.get_text()) if date_el else '',
+                    'league': _cl(league_el.get_text()) if league_el else '',
+                    'home': _cl(home_el.get_text()) if home_el else '',
+                    'away': _cl(away_el.get_text()) if away_el else '',
+                    'score': _cl(score_el.get_text()) if score_el else 'vs',
+                })
+        if recent_matches:
+            result['recent_matches'] = recent_matches
+            result['sections'].append('recent')
+
+    return result