#!/usr/bin/env python3
from dotenv import load_dotenv
load_dotenv('/opt/rinet-gpu/.env.master')
# auto-added by patch_scrapers_with_dotenv.sh
# ═══════════════════════════════════════════════════════════════════
# Fajl: sport_rezultati_arhivar.py | v1.0.0 | 05.05.2026
# Lokacija: /opt/pgz-sport/scrapers/sport_rezultati_arhivar.py
# Svrha: Wikipedia HR sezone HNL + Kup HR po godinama
#   - Iterate kroz sve sezone HNL od 1992
#   - Wikipedia API pages: "1._HNL_2023/24", "Kup_Hrvatske_u_nogometu_2024/25"
#   - Extract konacne tablice + finalne utakmice
#   - Plus PGŽ klubovi: HNK Rijeka, Orijent, Crikvenica, Opatija
# ═══════════════════════════════════════════════════════════════════
"""Sport rezultati historical arhivar."""
import os, re, json, time, hashlib
import urllib.request, urllib.parse
import psycopg2
from psycopg2.extras import execute_batch

DSN = f"host=10.10.0.2 port=6432 dbname=rinet_v3 user=rinet password={os.environ['DB_PASSWORD']}"
UA = "Ri.NET Civic Bot 1.0 (contact: dradulic@outlook.com)"
API = "https://hr.wikipedia.org/w/api.php"


def wiki_extract(title, sentences=None):
    params = {"action": "query", "prop": "extracts", "explaintext": "1",
              "redirects": "1", "format": "json", "titles": title}
    if sentences:
        params["exsentences"] = str(sentences)
    url = API + "?" + urllib.parse.urlencode(params)
    req = urllib.request.Request(url, headers={"User-Agent": UA})
    try:
        with urllib.request.urlopen(req, timeout=15) as r:
            d = json.loads(r.read())
        for pid, p in d.get("query", {}).get("pages", {}).items():
            if pid == "-1":
                return None
            return p.get("extract", "")
    except Exception as e:
        return None


def chunk(text, max_len=700):
    if len(text) <= max_len:
        return [text] if text else []
    out = []; start = 0
    while start < len(text):
        end = min(start + max_len, len(text))
        if end < len(text):
            for sep in [". ", "! ", "? ", "\n"]:
                p = text.rfind(sep, start, end)
                if p > start + max_len // 2:
                    end = p + len(sep); break
        out.append(text[start:end].strip())
        start = end
    return [c for c in out if len(c) > 80]


def insert_facts(conn, page, text, category, confidence=0.88):
    if not text or len(text) < 200:
        return 0
    cur = conn.cursor()
    rows = []
    for c in chunk(text, 700):
        h = hashlib.md5(c.encode()).hexdigest()
        rows.append((c, "wikipedia_sport_arhiv", category, confidence, h,
                     json.dumps({"page": page})))
    sql = """INSERT INTO dabi.knowledge (fact, source, category, confidence, data_hash, source_refs)
             VALUES (%s, %s, %s, %s, %s, %s::jsonb) ON CONFLICT (data_hash) DO NOTHING"""
    try:
        execute_batch(cur, sql, rows, page_size=50)
        n = cur.rowcount; cur.close()
        return n
    except Exception as e:
        return 0


def main():
    conn = psycopg2.connect(DSN); conn.autocommit = True
    
    pages = []
    
    # 1. HNL sezone 1992-2024
    for year in range(1992, 2026):
        for fmt in [f"1._HNL_{year}.", f"1._HNL_{year}./{(year+1)%100:02d}.",
                    f"HNL_{year}/{(year+1)%100:02d}", f"HNL_{year}-{year+1}",
                    f"SuperSport_HNL_{year}./{(year+1)%100:02d}.",
                    f"HT_Prva_HNL_{year}./{(year+1)%100:02d}."]:
            pages.append(("hnl_sezona", fmt))
    
    # 2. Kup Hrvatske u nogometu (po godinama)
    for year in range(1992, 2026):
        for fmt in [f"Kup_Hrvatske_u_nogometu_{year}.",
                    f"Kup_Hrvatske_u_nogometu_{year}./{(year+1)%100:02d}.",
                    f"Hrvatski_nogometni_kup_{year}-{year+1}"]:
            pages.append(("hr_nogometni_kup", fmt))
    
    # 3. Glavni klubovi PGŽ + povijest
    for klub in ["HNK_Rijeka", "NK_Orijent", "NK_Krk", "NK_Crikvenica",
                 "NK_Opatija", "NK_Mat-Promet", "NK_Pomorac", "NK_Naša_Slatina",
                 "HNK_Rijeka_(boys)", "ŽNK_Rijeka",
                 "HKK_Kvarner", "KK_Kvarner_2010", "KK_Lovran",
                 "HMRK_Zamet", "MRK_Pomorac", "RK_Trsat", "RK_Crikvenica",
                 "VK_Primorje", "VK_Rijeka",
                 "HRK_Rijeka", "HOK_Rijeka", "OK_Rijeka",
                 "HAOK_Mladost", "HAOK_Rijeka"]:
        pages.append(("pgz_klub_povijest", klub))
    
    # 4. Sezone HNK Rijeka po godinama
    for year in range(1990, 2026):
        for fmt in [f"Sezona_HNK_Rijeka_{year}./{(year+1)%100:02d}.",
                    f"HNK_Rijeka_u_sezoni_{year}-{year+1}",
                    f"HNK_Rijeka_{year}-{year+1}_sezona"]:
            pages.append(("hnk_rijeka_sezona", fmt))
    
    # Crawl
    successful = 0
    total_facts = 0
    found_pages = []
    
    for category, page in pages:
        text = wiki_extract(page)
        if text and len(text) > 300:
            successful += 1
            facts_inserted = insert_facts(conn, page, text, category, confidence=0.88)
            total_facts += facts_inserted
            found_pages.append(page)
            if successful % 10 == 0:
                print(f"  progress: {successful} pages found, {total_facts} facts")
        time.sleep(0.4)  # rate limit
    
    print(f"\n=== DONE: {successful}/{len(pages)} pages found, {total_facts} facts ===")
    print(f"Sample found pages: {found_pages[:15]}")
    conn.close()
    return {"pages_found": successful, "pages_tried": len(pages),
            "facts": total_facts}


if __name__ == "__main__":
    print(json.dumps(main()))