#!/usr/bin/env python3
"""JLS PGZ — 36 jedinica."""
import sys, json, time
sys.path.insert(0, "/opt/pgz-sport/scrapers/harvesters")
from _common import (fetch, extract_text, extract_title, chunk_text,
                      upsert_facts, find_internal_links, DSN)
from urllib.parse import urlparse
import psycopg2

JLS_PGZ = {
    "Rijeka":          "https://www.rijeka.hr/",
    "Opatija":         "https://www.opatija.hr/",
    "Crikvenica":      "https://www.crikvenica.hr/",
    "Krk":             "https://www.grad-krk.hr/",
    "Kraljevica":      "https://www.kraljevica.hr/",
    "Rab":             "https://www.rab.hr/",
    "Cres":            "https://www.cres.hr/",
    "Mali_Losinj":     "https://www.mali-losinj.hr/",
    "Delnice":         "https://www.delnice.hr/",
    "Vrbovsko":        "https://www.vrbovsko.hr/",
    "Cabar":           "https://www.cabar.hr/",
    "Bakar":           "https://www.bakar.hr/",
    "Kastav":          "https://www.kastav.hr/",
    "Novi_Vinodolski": "https://www.novi-vinodolski.hr/",
    "Viskovo":         "https://www.opcina-viskovo.hr/",
    "Klana":           "https://www.klana.hr/",
    "Moscenicka_Draga":"https://www.moscenicka-draga.hr/",
    "Lovran":          "https://www.opcinalovran.hr/",
    "Matulji":         "https://www.matulji.hr/",
    "Omisalj":         "https://www.omisalj.hr/",
    "Punat":           "https://www.punat.hr/",
    "Vrbnik":          "https://www.vrbnik.hr/",
    "Baska":           "https://www.baska.hr/",
    "Dobrinj":         "https://www.opcina-dobrinj.hr/",
    "Malinska":        "https://www.malinska.hr/",
    "Jelenje":         "https://www.jelenje.hr/",
    "Kostrena":        "https://www.kostrena.hr/",
    "Cavle":           "https://www.cavle.hr/",
    "Lopar":           "https://www.opcina-lopar.hr/",
    "Brod_Moravice":   "https://www.brod-moravice.hr/",
    "Mrkopalj":        "https://www.mrkopalj.hr/",
    "Ravna_Gora":      "https://www.ravnagora.hr/",
    "Lokve":           "https://www.lokve.hr/",
    "Skrad":           "https://www.skrad.hr/",
    "Fuzine":          "https://www.fuzine.hr/",
    "Vinodolska":      "https://www.vinodolska-opcina.hr/",
    "PGZ_zupanija":    "https://www.pgz.hr/",
}


def crawl(name, root, max_pages=25):
    conn = psycopg2.connect(DSN); conn.autocommit = True
    visited = set(); queue = [root]; facts = 0
    base_host = urlparse(root).hostname or ""
    while queue and len(visited) < max_pages:
        url = queue.pop(0)
        if url in visited: continue
        visited.add(url)
        html, status = fetch(url, timeout=15)
        if not html or status != 200: continue
        title = extract_title(html); text = extract_text(html)
        if not text or len(text) < 200: continue
        ff = []
        if title and len(title) > 8:
            ff.append({"fact": f"{name} - {title}", "url": url, "title": title})
        for c in chunk_text(text, 800):
            if len(c) > 100:
                ff.append({"fact": c, "url": url, "title": title})
        facts += upsert_facts(conn, ff, source_name=f"jls_pgz_{name.lower()}",
                               category="jls_pgz_official", confidence=0.90)
        for link in find_internal_links(html, url):
            if link not in visited and (urlparse(link).hostname or "") == base_host and len(queue) < 200:
                queue.append(link)
        time.sleep(0.4)
    conn.close()
    return {"name": name, "visited": len(visited), "facts": facts}


def main():
    results = []
    for name, url in JLS_PGZ.items():
        try:
            r = crawl(name, url, max_pages=25)
            results.append(r)
            print(f"  {name:25} {r['visited']:>3}p  {r['facts']:>5}f")
        except Exception as e:
            print(f"  {name:25} FAIL: {str(e)[:60]}")
    total = sum(r.get("facts", 0) for r in results)
    print(f"=== TOTAL: {total} ===")
    print(json.dumps({"jls_count": len(results), "total_facts": total}))


if __name__ == "__main__":
    main()