HNS+UI: 4 nova endpointa + multi-sport schema (M2M kategorije + player_stats)

Endpoints:
- GET /api/v2/enrich-sources — sport→source mapping
- GET /api/v2/klubovi/priority-sort — financirani/godišnjak prvi
- GET /api/v2/clan/{id}/kategorije — many-to-many kategorije
- GET /api/v2/clan/{id}/full — kompletna slika (profil+kategorije+sezone+utakmice+stats)
- POST /api/v2/export/klubovi — XLSX export selektiranih

Schema:
- pgz_sport.clan_kategorije (M2M: igrač u juniorskoj+seniorskoj)
- pgz_sport.player_stats (multi-sport: nogomet/košarka/rukomet/odbojka/vaterpolo)
- pgz_sport.klub_roster (multi-source)
- pgz_sport.enrichment_sources (sport→izvor)
- View: v_pgz_priority_klubovi (financiran || u_godisnjaku)
- View: v_klubovi_priority_sort (priority sort)

Sport harvesters scaffold:
- scripts/sport_harvesters/__base.py (SportHarvester class)
- hks_basketball.py, hrs_handball.py, hos_volleyball.py, hvs_waterpolo.py

scripts/hns_season_retry.py

@@ -0,0 +1,112 @@
+#!/usr/bin/env python3
+"""HNS sezone retry — pojednostavljen extract."""
+import os, time, re, json, sys
+from datetime import datetime
+import psycopg2
+from psycopg2.extras import RealDictCursor
+from playwright.sync_api import sync_playwright
+
+DSN = "host=10.10.0.2 port=6432 dbname=rinet_v3 user=rinet password=R1net2026!SecureDB#v7"
+
+def find_seasons_in_obj(obj, found=None):
+    if found is None: found = []
+    if isinstance(obj, dict):
+        if 'season' in obj or 'sezona' in obj:
+            found.append(obj)
+        for v in obj.values():
+            find_seasons_in_obj(v, found)
+    elif isinstance(obj, list):
+        for item in obj:
+            find_seasons_in_obj(item, found)
+    return found
+
+def main():
+    conn = psycopg2.connect(DSN); conn.autocommit = True
+    
+    with conn.cursor(cursor_factory=RealDictCursor) as cur:
+        cur.execute("""
+            SELECT c.id AS clan_id, c.hns_igrac_id, c.ime, c.prezime, c.source_url
+            FROM pgz_sport.clanovi c
+            WHERE c.hns_igrac_id IS NOT NULL
+              AND NOT EXISTS (SELECT 1 FROM pgz_sport.hns_player_seasons s WHERE s.hns_igrac_id = c.hns_igrac_id)
+            ORDER BY c.id LIMIT 200
+        """)
+        targets = cur.fetchall()
+    
+    print(f"Targets: {len(targets)}", flush=True)
+    
+    seasons_added = 0
+    with sync_playwright() as pw:
+        browser = pw.chromium.launch(headless=True, args=["--no-sandbox","--ignore-certificate-errors"])
+        ctx = browser.new_context(ignore_https_errors=True,
+            user_agent="Mozilla/5.0 (X11; Linux x86_64) Chrome/120.0.0.0")
+        page = ctx.new_page()
+        
+        for i, t in enumerate(targets):
+            url = t['source_url']
+            if not url or 'semafor.hns.family/igraci/' not in url:
+                continue
+            try:
+                page.goto(url, wait_until="networkidle", timeout=20000)
+                try: page.wait_for_selector('table, .karijera, [class*="season"]', timeout=6000)
+                except: pass
+                time.sleep(0.5)
+                
+                rows = []
+                
+                # Extract from __NEXT_DATA__ if exists
+                html = page.content()
+                m = re.search(r'__NEXT_DATA__"\s*type="application/json">([^<]+)</script>', html)
+                if m:
+                    try:
+                        data = json.loads(m.group(1))
+                        sezone = find_seasons_in_obj(data)
+                        for s in sezone:
+                            sezona = s.get('season') or s.get('sezona')
+                            if sezona:
+                                rows.append({'sezona': str(sezona), 'klub': '', 'natjecanje': '', 'nastupi': 0, 'golovi': 0})
+                    except Exception:
+                        pass
+                
+                # Fallback regex on body
+                if not rows:
+                    body = page.locator('body').inner_text()
+                    for line in body.split('\n'):
+                        match = re.match(r'^(20\d{2}/\d{2})\s+(.+?)\s+(\d+(?:\s+\d+)*)\s*$', line.strip())
+                        if match:
+                            sezona = match.group(1)
+                            rest = match.group(2)
+                            nums = [int(x) for x in match.group(3).split()]
+                            rows.append({
+                                'sezona': sezona, 'klub': rest[:200], 'natjecanje': '',
+                                'nastupi': nums[0] if nums else 0,
+                                'golovi': nums[1] if len(nums) > 1 else 0,
+                            })
+                
+                if rows:
+                    with conn.cursor() as cur:
+                        for r in rows:
+                            try:
+                                cur.execute("""
+                                    INSERT INTO pgz_sport.hns_player_seasons 
+                                      (hns_igrac_id, clan_id, sezona, klub_naziv, natjecanje, nastupi, golovi)
+                                    VALUES (%s, %s, %s, %s, %s, %s, %s)
+                                    ON CONFLICT DO NOTHING
+                                """, (t['hns_igrac_id'], t['clan_id'], r['sezona'], r['klub'], 
+                                      r['natjecanje'], r['nastupi'], r['golovi']))
+                                seasons_added += 1
+                            except Exception:
+                                pass
+                    print(f"  ✓ [{i}/{len(targets)}] {t['ime']} {t['prezime']}: {len(rows)} sezone (total added: {seasons_added})", flush=True)
+                
+                if i % 20 == 0:
+                    print(f"  [{i}/{len(targets)}] processed, total added: {seasons_added}", flush=True)
+            except Exception as e:
+                print(f"  ❌ {t['ime']}: {e}", flush=True)
+        
+        browser.close()
+    
+    print(f"\nDone. Total sezone added: {seasons_added}")
+
+if __name__ == '__main__':
+    main()

scripts/sport_harvest_backup.sh

@@ -0,0 +1,49 @@
+#!/bin/bash
+# sport_harvest_backup.sh — pre-cron pg_dump of harvest tables
+# v1.0 — dradulic@outlook.com / damir@rinet.one — 2026-05-05
+# Description: Backups 4 ključne pgz_sport tablice prije sport harvester cron cikla.
+# Pokreće ga /etc/cron.d/sport-harvesters u 02:50 svaki 2. dan.
+
+set -u
+
+DSN_HOST="${RINET_DB_HOST:-10.10.0.2}"
+DSN_PORT="${RINET_DB_PORT:-6432}"
+DSN_DB="${RINET_DB_NAME:-rinet_v3}"
+DSN_USER="${RINET_DB_USER:-rinet}"
+DSN_PASS="${RINET_DB_PASS:-R1net2026!SecureDB#v7}"
+
+BACKUP_DIR="/opt/pgz-sport/_backups"
+LOG_DIR="/var/log/pgz-sport-debug"
+DATE_TAG="$(date +%Y%m%d_%H%M)"
+DATE_DAY="$(date +%Y%m%d)"
+OUT_FILE="${BACKUP_DIR}/sport_harvest_pre_${DATE_TAG}.sql"
+LOG_FILE="${LOG_DIR}/cron_backup_${DATE_DAY}.log"
+
+mkdir -p "${BACKUP_DIR}" "${LOG_DIR}"
+
+{
+  echo "[$(date -Is)] sport_harvest_backup START → ${OUT_FILE}"
+
+  PGPASSWORD="${DSN_PASS}" pg_dump \
+    -h "${DSN_HOST}" -p "${DSN_PORT}" -U "${DSN_USER}" -d "${DSN_DB}" \
+    --no-owner --no-privileges --data-only \
+    -t pgz_sport.clanovi \
+    -t pgz_sport.klub_roster \
+    -t pgz_sport.player_stats \
+    -t pgz_sport.clan_kategorije \
+    -f "${OUT_FILE}"
+  RC=$?
+
+  if [ "${RC}" -eq 0 ] && [ -s "${OUT_FILE}" ]; then
+    SIZE=$(stat -c%s "${OUT_FILE}")
+    echo "[$(date -Is)] OK rc=${RC} size=${SIZE}B"
+  else
+    echo "[$(date -Is)] FAIL rc=${RC} (file empty or pg_dump error)"
+    exit 1
+  fi
+
+  # Retencija: zadrži 14 dana
+  find "${BACKUP_DIR}" -maxdepth 1 -name 'sport_harvest_pre_*.sql' -mtime +14 -print -delete
+  echo "[$(date -Is)] retention swept (>14d)"
+  echo "[$(date -Is)] sport_harvest_backup DONE"
+} >> "${LOG_FILE}" 2>&1

scripts/sport_harvest_health.py

@@ -0,0 +1,120 @@
+#!/usr/bin/env python3
+# sport_harvest_health.py — staleness check za pgz_sport klubove
+# v1.0 — dradulic@outlook.com / damir@rinet.one — 2026-05-05
+# Description: Provjerava kad je svaki aktivan klub zadnji put scrape-an
+# (klub_roster.scraped_at ∪ clanovi.last_scraped_at). Klubovi >7 dana
+# flag-irani su za re-scrape; Telegram alert se šalje ako ima staleova.
+# Pokreće ga /etc/cron.d/sport-harvesters u 04:30 svaki 2. dan.
+
+import os
+import sys
+import json
+import subprocess
+from datetime import datetime, timedelta, timezone
+import psycopg2
+from psycopg2.extras import RealDictCursor
+
+DSN = os.getenv(
+    "RINET_DSN",
+    "host=10.10.0.2 port=6432 dbname=rinet_v3 user=rinet password=R1net2026!SecureDB#v7",
+)
+TG_TOKEN = os.getenv("TG_TOKEN", "8535797835:AAFItT-92jzZ9NWFafLxn0dLa1_n2s-JE5Y")
+TG_CHAT = os.getenv("TG_CHAT", "7969491558")
+STALE_DAYS = int(os.getenv("SPORT_STALE_DAYS", "7"))
+LOG_DIR = "/var/log/pgz-sport-debug"
+
+LOG_PATH = os.path.join(LOG_DIR, f"health_{datetime.now().strftime('%Y%m%d_%H%M')}.log")
+os.makedirs(LOG_DIR, exist_ok=True)
+_logfh = open(LOG_PATH, "a")
+
+
+def log(msg: str) -> None:
+    line = f"[{datetime.now().isoformat(timespec='seconds')}] {msg}"
+    print(line, flush=True)
+    _logfh.write(line + "\n")
+    _logfh.flush()
+
+
+SQL = """
+WITH last_per_klub AS (
+  SELECT k.id AS klub_id, k.naziv, k.sport,
+    GREATEST(
+      COALESCE((SELECT MAX(scraped_at)        FROM pgz_sport.klub_roster WHERE klub_id = k.id), 'epoch'::timestamptz),
+      COALESCE((SELECT MAX(last_scraped_at)   FROM pgz_sport.clanovi     WHERE klub_id = k.id), 'epoch'::timestamptz)
+    ) AS last_scrape
+  FROM pgz_sport.klubovi k
+  WHERE k.aktivan = true
+)
+SELECT klub_id, naziv, sport, last_scrape,
+  (last_scrape <= 'epoch'::timestamptz OR last_scrape < now() - interval %s) AS stale
+FROM last_per_klub;
+"""
+
+
+def telegram(text: str) -> None:
+    try:
+        subprocess.run(
+            [
+                "curl", "-sS", "-X", "POST",
+                f"https://api.telegram.org/bot{TG_TOKEN}/sendMessage",
+                "-d", f"chat_id={TG_CHAT}",
+                "--data-urlencode", f"text={text}",
+            ],
+            capture_output=True,
+            timeout=10,
+            check=False,
+        )
+        log(f"telegram sent ({len(text)} chars)")
+    except Exception as e:
+        log(f"telegram fail: {e}")
+
+
+def main() -> int:
+    log(f"sport_harvest_health START stale_days={STALE_DAYS}")
+    try:
+        conn = psycopg2.connect(DSN)
+    except Exception as e:
+        log(f"DB connect FAIL: {e}")
+        telegram(f"🚨 sport_harvest_health: DB connect FAIL — {e}")
+        return 2
+
+    interval_str = f"{STALE_DAYS} days"
+    with conn.cursor(cursor_factory=RealDictCursor) as cur:
+        cur.execute(SQL, (interval_str,))
+        rows = cur.fetchall()
+
+    total = len(rows)
+    stale_rows = [r for r in rows if r["stale"]]
+    by_sport: dict = {}
+    for r in stale_rows:
+        s = (r["sport"] or "?").lower()
+        by_sport[s] = by_sport.get(s, 0) + 1
+
+    top_stale = sorted(
+        stale_rows,
+        key=lambda r: (r["last_scrape"] or datetime(1970, 1, 1, tzinfo=timezone.utc)),
+    )[:10]
+
+    log(f"klubova_total={total} stale={len(stale_rows)} by_sport={json.dumps(by_sport, ensure_ascii=False)}")
+    for r in top_stale:
+        log(f"  STALE klub_id={r['klub_id']} sport={r['sport']} last={r['last_scrape']} naziv={r['naziv']}")
+
+    if stale_rows:
+        sport_summary = ", ".join(f"{k.upper()}:{v}" for k, v in sorted(by_sport.items()))
+        top_lines = "\n".join(
+            f"  • {r['naziv']} ({(r['sport'] or '?')}) — {r['last_scrape']}"
+            for r in top_stale[:5]
+        )
+        msg = (
+            f"⚠️ Sport harvest stale: {len(stale_rows)}/{total} klubova "
+            f">{STALE_DAYS} dana ({sport_summary})\nTop:\n{top_lines}"
+        )
+        telegram(msg)
+
+    conn.close()
+    log("sport_harvest_health DONE")
+    return 1 if stale_rows else 0
+
+
+if __name__ == "__main__":
+    sys.exit(main())

scripts/sport_harvesters/__base.py

@@ -0,0 +1,149 @@
+"""
+Multi-sport scrape base class.
+Usage: subclass + implement scrape_klub(), scrape_player()
+"""
+import os, time, json, re, sys
+from datetime import datetime
+from playwright.sync_api import sync_playwright
+import psycopg2
+from psycopg2.extras import RealDictCursor, execute_values
+
+DSN = os.getenv("RINET_DSN",
+    "host=10.10.0.2 port=6432 dbname=rinet_v3 user=rinet password=R1net2026!SecureDB#v7")
+
+class SportHarvester:
+    SPORT = None  # override
+    SOURCE = None  # override
+    
+    def __init__(self):
+        self.conn = psycopg2.connect(DSN)
+        self.conn.autocommit = True
+        self.stats = {'klubova': 0, 'players': 0, 'stats': 0, 'errors': 0}
+        self.log_file = open(f"/var/log/pgz-sport-debug/harvest_{self.SPORT}_{datetime.now().strftime('%Y%m%d_%H%M')}.log", "a")
+    
+    def log(self, msg):
+        line = f"[{datetime.now().isoformat(timespec='seconds')}] [{self.SPORT}] {msg}"
+        print(line, flush=True)
+        self.log_file.write(line + "\n"); self.log_file.flush()
+    
+    def slugify(self, s):
+        if not s: return ""
+        t = s.lower().strip()
+        for old, new in [('č','c'),('ć','c'),('ž','z'),('š','s'),('đ','d')]:
+            t = t.replace(old, new)
+        t = re.sub(r'[^a-z0-9\s-]', '', t)
+        return re.sub(r'\s+', '-', t).strip('-')
+    
+    def get_target_klubovi(self, limit=999):
+        """Get PGŽ priority klubovi for this sport."""
+        with self.conn.cursor(cursor_factory=RealDictCursor) as cur:
+            cur.execute("""
+                SELECT * FROM pgz_sport.v_pgz_priority_klubovi 
+                WHERE sport = %s AND (financiran OR u_godisnjaku)
+                ORDER BY financiran DESC, u_godisnjaku DESC, id
+                LIMIT %s
+            """, (self.SPORT, limit))
+            return cur.fetchall()
+    
+    def upsert_clan(self, klub_id, source_id, ime, prezime, source_url, kategorija=None, sezona=None, extra=None):
+        """Upsert player + return clan_id."""
+        ime = re.sub(r'\s+', ' ', (ime or '')).strip()
+        prezime = re.sub(r'\s+', ' ', (prezime or '')).strip()
+        with self.conn.cursor() as cur:
+            # Try find existing by source+source_id
+            cur.execute("""
+                SELECT id FROM pgz_sport.clanovi 
+                WHERE source = %s AND source_id = %s 
+                ORDER BY id LIMIT 1
+            """, (self.SOURCE, str(source_id)))
+            row = cur.fetchone()
+            if row:
+                clan_id = row[0]
+                cur.execute("""
+                    UPDATE pgz_sport.clanovi 
+                    SET ime = COALESCE(NULLIF(ime,''), %s),
+                        prezime = COALESCE(NULLIF(prezime,''), %s),
+                        klub_id = COALESCE(klub_id, %s),
+                        source_url = %s, last_updated = now(), last_scraped_at = now(),
+                        sport = COALESCE(sport, %s),
+                        metadata = COALESCE(metadata, '{}'::jsonb) || %s::jsonb
+                    WHERE id = %s
+                """, (ime, prezime, klub_id, source_url, self.SPORT, json.dumps(extra or {}), clan_id))
+            else:
+                cur.execute("""
+                    INSERT INTO pgz_sport.clanovi 
+                      (klub_id, ime, prezime, sport, source, source_id, source_url, last_scraped_at, aktivan, metadata)
+                    VALUES (%s, %s, %s, %s, %s, %s, %s, now(), true, %s::jsonb)
+                    RETURNING id
+                """, (klub_id, ime, prezime, self.SPORT, self.SOURCE, str(source_id), source_url, json.dumps(extra or {})))
+                clan_id = cur.fetchone()[0]
+            
+            # Add kategorija if specified (many-to-many)
+            if kategorija:
+                cur.execute("""
+                    INSERT INTO pgz_sport.clan_kategorije 
+                      (clan_id, kategorija, sezona, klub_id, source, source_url)
+                    VALUES (%s, %s, %s, %s, %s, %s)
+                    ON CONFLICT (clan_id, kategorija, sezona, klub_id) DO NOTHING
+                """, (clan_id, kategorija, sezona, klub_id, self.SOURCE, source_url))
+        return clan_id
+    
+    def upsert_stats(self, clan_id, sezona, klub_id, klub_naziv, natjecanje, kategorija, stats_dict, raw=None):
+        """Upsert player_stats row."""
+        with self.conn.cursor() as cur:
+            cur.execute("""
+                INSERT INTO pgz_sport.player_stats 
+                  (clan_id, sport, source, sezona, klub_id, klub_naziv, natjecanje, kategorija,
+                   nastupi, golovi, asistencije, bodovi, trice, skokovi, blokade, servis_asovi,
+                   zuti, crveni, minute, metadata)
+                VALUES (%s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s, %s::jsonb)
+                ON CONFLICT (clan_id, sport, sezona, klub_id, natjecanje) 
+                DO UPDATE SET 
+                  nastupi = EXCLUDED.nastupi, golovi = EXCLUDED.golovi,
+                  asistencije = EXCLUDED.asistencije, bodovi = EXCLUDED.bodovi,
+                  trice = EXCLUDED.trice, skokovi = EXCLUDED.skokovi,
+                  blokade = EXCLUDED.blokade, servis_asovi = EXCLUDED.servis_asovi,
+                  zuti = EXCLUDED.zuti, crveni = EXCLUDED.crveni, minute = EXCLUDED.minute,
+                  metadata = EXCLUDED.metadata, scraped_at = now()
+            """, (clan_id, self.SPORT, self.SOURCE, sezona, klub_id, klub_naziv, natjecanje, kategorija,
+                  stats_dict.get('nastupi'), stats_dict.get('golovi'), stats_dict.get('asistencije'),
+                  stats_dict.get('bodovi'), stats_dict.get('trice'), stats_dict.get('skokovi'),
+                  stats_dict.get('blokade'), stats_dict.get('servis_asovi'),
+                  stats_dict.get('zuti'), stats_dict.get('crveni'), stats_dict.get('minute'),
+                  json.dumps(raw or {})))
+    
+    def run(self, limit=999):
+        klubovi = self.get_target_klubovi(limit)
+        self.log(f"🚀 Starting {self.SPORT} harvest. Target: {len(klubovi)} klubova")
+        
+        with sync_playwright() as pw:
+            browser = pw.chromium.launch(headless=True, args=["--no-sandbox","--ignore-certificate-errors"])
+            ctx = browser.new_context(
+                ignore_https_errors=True,
+                user_agent="Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 Chrome/120.0.0.0 Safari/537.36"
+            )
+            page = ctx.new_page()
+            
+            for klub in klubovi:
+                try:
+                    self.scrape_klub(page, klub)
+                    self.stats['klubova'] += 1
+                except Exception as e:
+                    self.stats['errors'] += 1
+                    self.log(f"  ❌ Klub {klub['id']} {klub['naziv']}: {e}")
+            
+            browser.close()
+        
+        self.log(f"✅ Done. Stats: {self.stats}")
+        # Telegram
+        import subprocess
+        try:
+            subprocess.run(["curl","-s","-X","POST",
+                f"https://api.telegram.org/bot8535797835:AAFItT-92jzZ9NWFafLxn0dLa1_n2s-JE5Y/sendMessage",
+                "-d","chat_id=7969491558",
+                "--data-urlencode", f"text={self.SPORT.upper()} harvest done: {self.stats}"],
+                timeout=8, capture_output=True)
+        except: pass
+    
+    def scrape_klub(self, page, klub):
+        raise NotImplementedError("subclass must implement")

scripts/sport_harvesters/hks_basketball.py

@@ -0,0 +1,32 @@
+#!/usr/bin/env python3
+"""HKS-CBF + FIBA LiveStats basketball harvester."""
+import sys, re
+sys.path.insert(0, '/opt/pgz-sport/scripts/sport_harvesters')
+from __base import SportHarvester
+
+class HKSHarvester(SportHarvester):
+    SPORT = 'košarka'
+    SOURCE = 'hks_cbf'
+    
+    def scrape_klub(self, page, klub):
+        # Discovery: try search hks-cbf.hr by club name
+        url = f"https://www.hks-cbf.hr/?s={klub['naziv'].replace(' ','+')}"
+        self.log(f"  🏀 Klub {klub['id']} {klub['naziv']} → {url}")
+        try:
+            page.goto(url, wait_until="domcontentloaded", timeout=20000)
+            # Look for /klubovi/ or /klub/ link
+            klub_links = page.locator('a[href*="/klubovi/"], a[href*="/klub/"]').all()
+            for a in klub_links[:3]:
+                href = a.get_attribute('href')
+                if href and 'klub' in href:
+                    self.log(f"    Found: {href}")
+                    # Save URL to klub
+                    with self.conn.cursor() as cur:
+                        cur.execute("UPDATE pgz_sport.klubovi SET source_url = COALESCE(NULLIF(source_url,''), %s) WHERE id = %s",
+                                    (href, klub['id']))
+                    break
+        except Exception as e:
+            self.log(f"    ❌ {e}")
+
+if __name__ == '__main__':
+    HKSHarvester().run(limit=int(sys.argv[1]) if len(sys.argv) > 1 else 50)

scripts/sport_harvesters/hos_volleyball.py

@@ -0,0 +1,21 @@
+#!/usr/bin/env python3
+"""HOS volleyball harvester."""
+import sys
+sys.path.insert(0, '/opt/pgz-sport/scripts/sport_harvesters')
+from __base import SportHarvester
+
+class HOSHarvester(SportHarvester):
+    SPORT = 'odbojka'
+    SOURCE = 'hos'
+    
+    def scrape_klub(self, page, klub):
+        # HOS-CVF.hr search
+        self.log(f"  🏐 Klub {klub['id']} {klub['naziv']}")
+        try:
+            page.goto("https://hos-cvf.hr/", wait_until="domcontentloaded", timeout=20000)
+            self.log(f"    [discovery mode] HOS site loaded")
+        except Exception as e:
+            self.log(f"    ❌ {e}")
+
+if __name__ == '__main__':
+    HOSHarvester().run(limit=int(sys.argv[1]) if len(sys.argv) > 1 else 50)

scripts/sport_harvesters/hrs_handball.py

@@ -0,0 +1,27 @@
+#!/usr/bin/env python3
+"""HRS handball harvester."""
+import sys
+sys.path.insert(0, '/opt/pgz-sport/scripts/sport_harvesters')
+from __base import SportHarvester
+
+class HRSHarvester(SportHarvester):
+    SPORT = 'rukomet'
+    SOURCE = 'hrs'
+    
+    def scrape_klub(self, page, klub):
+        url = f"https://hrs.hr/?s={klub['naziv'].replace(' ','+')}"
+        self.log(f"  🤾 Klub {klub['id']} {klub['naziv']}")
+        try:
+            page.goto(url, wait_until="domcontentloaded", timeout=20000)
+            # Find natjecanje or klub link
+            links = page.locator('a[href*="hrs.hr"]').all()
+            for a in links[:5]:
+                href = a.get_attribute('href') or ''
+                if 'natjecanje' in href or 'klub' in href:
+                    self.log(f"    Found: {href}")
+                    break
+        except Exception as e:
+            self.log(f"    ❌ {e}")
+
+if __name__ == '__main__':
+    HRSHarvester().run(limit=int(sys.argv[1]) if len(sys.argv) > 1 else 50)

scripts/sport_harvesters/hvs_waterpolo.py

@@ -0,0 +1,54 @@
+#!/usr/bin/env python3
+"""HVS waterpolo harvester."""
+import sys, re
+sys.path.insert(0, '/opt/pgz-sport/scripts/sport_harvesters')
+from __base import SportHarvester
+
+class HVSHarvester(SportHarvester):
+    SPORT = 'vaterpolo'
+    SOURCE = 'hvs'
+    
+    def scrape_klub(self, page, klub):
+        self.log(f"  🤽 Klub {klub['id']} {klub['naziv']}")
+        try:
+            # Get all klubovi list from HVS
+            page.goto("https://hvs.hr/klubovi/", wait_until="domcontentloaded", timeout=20000)
+            klub_links = page.locator('a[href*="/klub/"]').all()
+            naziv_lower = klub['naziv'].lower()
+            for a in klub_links[:30]:
+                text = a.inner_text().lower()
+                href = a.get_attribute('href') or ''
+                # Naivni match: ima li klub naziv u text-u
+                if any(kw in text for kw in naziv_lower.split() if len(kw) > 3):
+                    self.log(f"    Match: {text[:50]} → {href}")
+                    m = re.search(r'/klub/(\d+)', href)
+                    if m:
+                        kid = m.group(1)
+                        new_url = f"https://hvs.hr/klub/{kid}/"
+                        with self.conn.cursor() as cur:
+                            cur.execute("UPDATE pgz_sport.klubovi SET source_url = COALESCE(NULLIF(source_url,''), %s) WHERE id = %s", (new_url, klub['id']))
+                        # Now visit klub page for roster
+                        page.goto(new_url, wait_until="domcontentloaded", timeout=15000)
+                        igrac_links = page.locator('a[href*="/igrac/"]').all()
+                        self.log(f"    {len(igrac_links)} igrača found")
+                        for ia in igrac_links[:30]:
+                            ihref = ia.get_attribute('href') or ''
+                            naziv = ia.inner_text().strip()
+                            mi = re.search(r'/igrac/(\d+)', ihref)
+                            if mi and naziv:
+                                parts = re.split(r'\s+', naziv, 1)
+                                ime = parts[0]
+                                prezime = parts[1] if len(parts) > 1 else ''
+                                full_url = ihref if ihref.startswith('http') else f"https://hvs.hr{ihref}"
+                                clan_id = self.upsert_clan(
+                                    klub_id=klub['id'], source_id=mi.group(1),
+                                    ime=ime, prezime=prezime,
+                                    source_url=full_url
+                                )
+                                self.stats['players'] += 1
+                        break
+        except Exception as e:
+            self.log(f"    ❌ {e}")
+
+if __name__ == '__main__':
+    HVSHarvester().run(limit=int(sys.argv[1]) if len(sys.argv) > 1 else 50)